ROADMAP_EXTENSION.md

Phase: v167.x — Qudit / Ququart / High-Dimensional Stabilizer Receipts

Status PLANNED

Source-Grounded Motivation Qudit stabilizer codes generalize qubit stabilizer codes to local dimension d > 2. Qutrit (d=3) and ququart (d=4) codes appear in the QEC release history as early as v1.5 and v1.6, establishing ququart stabilizer codes and ternary Golay codes as first-class objects. The v91.x arc introduced qudit measurement and syndrome dynamics. The v134.x arc introduced generalized qudit lattice field engines.

These earlier arcs operated without receipt-bound invariant discipline. v167.x retrofits that discipline onto the high-dimensional stabilizer surface. The purpose is not to re-implement qudit codes but to produce receipt-bound artifacts that declare: local dimension policy, Galois field GF(d) encoding policy, stabilizer generator canonical form, syndrome extraction policy, and equivalence class declaration.

High-dimensional stabilizers are adapters. They are not automatically superior to qubit codes. Any claim that a qudit code achieves lower overhead, higher threshold, or better geometry requires a benchmark receipt and a replay equivalence receipt before entering the QEC proof pipeline.

Arc Reinterpretation

"High-dimensional stabilizer codes are adapter hypotheses.

They are not accepted because they are elegant.
They are accepted only when replay equivalence
against the canonical qubit baseline is proven
under declared dimension policy and Galois field encoding."

Planned Releases

• v167.0 → QuditDimensionPolicyManifest
• v167.1 → QuditStabilizerGeneratorReceipt
• v167.2 → QuditSyndromeExtractionReceipt
• v167.3 → QuditEquivalenceClassReceipt
• v167.4 → QuditCodeParameterReceipt
• v167.5 → QuditReplayEquivalenceReceipt
• v167.6 → QuditBenchmarkBoundaryReceipt

Expected Modules

• src/qec/analysis/qudit_dimension_policy_manifest.py
• src/qec/analysis/qudit_stabilizer_generator_receipts.py
• src/qec/analysis/qudit_syndrome_extraction_receipts.py
• src/qec/analysis/qudit_equivalence_class_receipts.py
• src/qec/analysis/qudit_code_parameter_receipts.py
• src/qec/analysis/qudit_replay_equivalence_receipts.py
• src/qec/analysis/qudit_benchmark_boundary_receipts.py

Expected Artifacts

• QuditDimensionPolicyManifest
• QuditStabilizerGeneratorReceipt
• QuditSyndromeExtractionReceipt
• QuditEquivalenceClassReceipt
• QuditCodeParameterReceipt
• QuditReplayEquivalenceReceipt
• QuditBenchmarkBoundaryReceipt

Expected Hashes

• qudit_dimension_policy_manifest_hash (v167.0)
• qudit_stabilizer_generator_receipt_hash (v167.1)
• qudit_syndrome_extraction_receipt_hash (v167.2)
• qudit_equivalence_class_receipt_hash (v167.3)
• qudit_code_parameter_receipt_hash (v167.4)
• qudit_replay_equivalence_receipt_hash (v167.5)
• qudit_benchmark_boundary_receipt_hash (v167.6)

Core Rule

same local dimension d
+ same Galois field GF(d) encoding policy
+ same stabilizer generator canonical form
+ same syndrome extraction policy
+ same input syndrome corpus
+ same equivalence class declaration
→ same canonical qudit stabilizer output
→ same qudit_replay_equivalence_receipt_hash

Acceptance Gates

• pytest: QuditDimensionPolicyManifest declares local dimension d, field GF(d), and canonical form policy
• pytest: QuditStabilizerGeneratorReceipt is hash-stable across PYTHONHASHSEED values
• pytest: QuditSyndromeExtractionReceipt declares measurement basis, extraction ordering, and output dtype
• pytest: QuditEquivalenceClassReceipt rejects equivalence claims without declared comparison mode
• pytest: QuditCodeParameterReceipt declares [[n,k,d]] parameters and distance verification method
• pytest: QuditReplayEquivalenceReceipt requires comparison against canonical qubit baseline for d=2 subcase
• pytest: QuditBenchmarkBoundaryReceipt rejects any overhead or threshold claim without benchmark receipt
• pytest: no live network calls in tests
• pytest: no hardware authority claims
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: qudit backends are adapters, never authorities

Must Not Do

• no qudit code accepted as superior without replay equivalence receipt
• no overhead claim without benchmark receipt
• no threshold claim without source-bound evidence
• no silent Galois field encoding
• no undeclared local dimension
• no hardware authority claim
• no cosmological geometry claim without claim-boundary receipt
• no qudit syndrome accepted without extraction policy declaration

Dependency Boundaries

• depends on: v166.x DecoderPromotionReceipt (canonical decoder baseline discipline)
• depends on: v163.x HeavyDependencyDiscoveryReceipt (backend invariant discipline)
• feeds into: v174.x BP Dynamics / Fixed-Point Trap Receipts (qudit BP dynamics)
• feeds into: v179.x Quantum Geometry / Contextuality receipts (qudit geometry signals)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v167.x extends the decoder governance discipline of v166.x into the high-dimensional stabilizer space. It does not reimplement the v1.5/v1.6/v91.x/v134.x qudit engines. It wraps their behavioral contracts as receipt-bound invariants, consistent with the adapter-not-authority law established in v163.x.

---

Phase: v168.x — Proof Telemetry / MIDI / Sonification Receipts

Status PLANNED

Source-Grounded Motivation The QEC release history contains multiple sonification layers: v82.3.0 introduced MIDI/cube/DNA bridge pipelines, v96.1.0 introduced deterministic sonification of hierarchical correction dynamics, v136.8.8 introduced a SID-inspired deterministic sonification backend, and v137.7.2 introduced hash-preserving compression chains with sonification projection.

These earlier arcs produced audio event sequences and PCM render specifications. However, they did not produce receipt-bound proof artifacts that could enter the canonical hash chain. v168.x formalizes the sonification surface as a proof telemetry layer.

Proof telemetry means: every proof event that QEC produces may be projected into a deterministic audio event sequence. The projection is a bounded, receipt-bound adapter. The audio output is not the proof. The receipt is the boundary.

MIDI sequences and PCM render specifications are deterministic artifacts when their generation policy is declared. They are not musical compositions. They are structured event streams derived from canonical proof state.

Arc Reinterpretation

"Sonification is not music.

Sonification is a deterministic projection
of canonical proof state into a structured event stream.

The event stream is a receipt.
The audio is not the proof.
The receipt is the boundary."

Planned Releases

• v168.0 → ProofTelemetryManifest
• v168.1 → MIDIEventStreamReceipt
• v168.2 → PCMRenderSpecReceipt
• v168.3 → SonificationProjectionReceipt
• v168.4 → TelemetryReplayEquivalenceReceipt

Expected Modules

• src/qec/analysis/proof_telemetry_manifest.py
• src/qec/analysis/midi_event_stream_receipts.py
• src/qec/analysis/pcm_render_spec_receipts.py
• src/qec/analysis/sonification_projection_receipts.py
• src/qec/analysis/telemetry_replay_equivalence_receipts.py

Expected Artifacts

• ProofTelemetryManifest
• MIDIEventStreamReceipt
• PCMRenderSpecReceipt
• SonificationProjectionReceipt
• TelemetryReplayEquivalenceReceipt

Expected Hashes

• proof_telemetry_manifest_hash (v168.0)
• midi_event_stream_receipt_hash (v168.1)
• pcm_render_spec_receipt_hash (v168.2)
• sonification_projection_receipt_hash (v168.3)
• telemetry_replay_equivalence_receipt_hash (v168.4)

Core Rule

same canonical proof state
+ same sonification projection policy
+ same MIDI event generation policy
+ same PCM render specification
+ same deterministic ordering
→ same structured event stream
→ same sonification_projection_receipt_hash

Acceptance Gates

• pytest: ProofTelemetryManifest declares proof state source, projection policy, and output format
• pytest: MIDIEventStreamReceipt is hash-stable across PYTHONHASHSEED values
• pytest: PCMRenderSpecReceipt declares sample rate, bit depth, channel count, and render policy
• pytest: SonificationProjectionReceipt declares mapping from proof state fields to audio parameters
• pytest: TelemetryReplayEquivalenceReceipt verifies same proof state produces same event stream
• pytest: no live audio rendering in tests (render spec only, not rendered output)
• pytest: no hardware audio device dependency in tests
• pytest: no nondeterministic timing in event stream generation
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: audio backends are adapters, never authorities

Must Not Do

• no audio output treated as proof
• no nondeterministic timing in event stream
• no hardware audio device dependency in tests
• no undeclared projection policy
• no MIDI sequence treated as canonical proof artifact
• no PCM render treated as replay-equivalent substitute for canonical JSON

Dependency Boundaries

• depends on: v166.x decoder promotion receipts (proof state source)
• depends on: v165.x optimized simulation reports (telemetry source)
• feeds into: v175.x Operator Console Unification (telemetry display surface)
• feeds into: v188.x Deterministic Experiment Scheduler (telemetry scheduling)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v168.x formalizes the sonification discipline established in v82.3.0, v96.1.0, v136.8.8, and v137.7.2. It does not replace those earlier layers. It wraps their behavioral contracts as receipt-bound proof telemetry artifacts, consistent with the adapter-not-authority law.

---

Phase: v169.x — Symbolic Geometry Grammar / Cosmovirus Sandbox

Status PLANNED

Source-Grounded Motivation The QEC repository contains qec_theory_diagram.txt as a symbolic grammar reference. The roadmap extension materials reference cosmovirus as a symbolic sandbox concept. The v137.10.x arc introduced hypothesis lattices and claim audit kernels. The v137.8.0 arc introduced a topological graph kernel.

Symbolic geometry grammar means: a bounded, receipt-bound system for expressing mathematical identity candidates, geometric claims, and cosmological signals. The sandbox is not a theorem prover. It is a claim-boundary surface. Every symbolic expression must declare: claim mode, evidence scope, review status, and sigma threshold (for cosmological signals).

The Cosmovirus Sandbox is not a metaphysical engine. It is a bounded symbolic claim registry. Claims that exceed declared evidence scope are rejected. Claims that lack proof receipts are labelled UNPROVEN. Claims that reference cosmological signals require sigma threshold gates and source-bound evidence.

Arc Reinterpretation

"Symbolic geometry grammar is a bounded claim registry,
not a theorem prover.

A symbolic hypothesis is not a proof.
A mathematical identity candidate is not a theorem.
A cosmological signal is not a truth claim.

The receipt is the boundary.
The claim_mode = SYMBOLIC_ONLY gate is mandatory."

Planned Releases

• v169.0 → SymbolicGrammarManifest
• v169.1 → SymbolicHypothesisReceipt
• v169.2 → MathematicalIdentityCandidateReceipt
• v169.3 → CosmologicalClaimBoundaryReceipt
• v169.4 → SymbolicSandboxReplayReceipt

Expected Modules

• src/qec/analysis/symbolic_grammar_manifest.py
• src/qec/analysis/symbolic_hypothesis_receipts.py
• src/qec/analysis/mathematical_identity_candidate_receipts.py
• src/qec/analysis/cosmological_claim_boundary_receipts.py
• src/qec/analysis/symbolic_sandbox_replay_receipts.py

Expected Artifacts

• SymbolicGrammarManifest
• SymbolicHypothesisReceipt
• MathematicalIdentityCandidateReceipt
• CosmologicalClaimBoundaryReceipt
• SymbolicSandboxReplayReceipt

Expected Hashes

• symbolic_grammar_manifest_hash (v169.0)
• symbolic_hypothesis_receipt_hash (v169.1)
• mathematical_identity_candidate_receipt_hash (v169.2)
• cosmological_claim_boundary_receipt_hash (v169.3)
• symbolic_sandbox_replay_receipt_hash (v169.4)

Core Rule

same symbolic grammar version
+ same claim_mode declaration
+ same evidence scope declaration
+ same review status declaration
+ same sigma threshold (for cosmological signals)
→ same symbolic claim output
→ same symbolic_sandbox_replay_receipt_hash

Acceptance Gates

• pytest: SymbolicGrammarManifest declares grammar version, claim_mode, and evidence scope
• pytest: SymbolicHypothesisReceipt is labelled UNPROVEN unless proof receipt exists
• pytest: MathematicalIdentityCandidateReceipt rejects theorem claims without proof receipt
• pytest: CosmologicalClaimBoundaryReceipt requires sigma threshold declaration and source-bound evidence
• pytest: SymbolicSandboxReplayReceipt verifies same symbolic input produces same canonical output
• pytest: claim_mode = SYMBOLIC_ONLY is enforced for all symbolic surfaces
• pytest: no numerology accepted as proof
• pytest: no model-generated theorem accepted as authority
• pytest: no cosmological signal without sigma threshold gate
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: symbolic backends are adapters, never authorities

Must Not Do

• no theorem claim without proof receipt
• no numerology as proof
• no model-generated theorem authority
• no cosmological truth claim
• no claim scope expansion beyond declared evidence
• no sigma threshold omission for cosmological signals
• no symbolic expression without claim_mode declaration
• no undeclared review status

Dependency Boundaries

• depends on: v165.6.x AI-Scientist provenance receipts (claim scope discipline)
• depends on: v165.9.x quantum geometry signal receipts (cosmological signal discipline)
• feeds into: v179.x Quantum Geometry / Contextuality receipts (symbolic geometry signals)
• feeds into: v183.x Reproducible Research Publication Receipts (symbolic claim provenance)
• feeds into: v186.x Symbolic Diagram Compiler v2 (grammar extension)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v169.x extends the claim-audit discipline of v137.10.x into the symbolic geometry and cosmological signal domain. It enforces the Mathematical Discovery Reminder already present in the roadmap: no theorem claim without proof receipt, no numerology as proof, no model-generated theorem authority.

---

Phase: v170.x — Reproducible Build / Supply-Chain Receipts

Status PLANNED

Source-Grounded Motivation The roadmap already establishes installer provenance hardening as a post-v161 hygiene priority. The Debian Forky mandatory reproducibility milestone is cited as a research basis. The risk register identifies installer supply-chain risk, license ambiguity, and version drift as explicit risks.

v170.x formalizes the reproducible build and supply-chain receipt discipline. A reproducible build means: same source inputs, same build environment declaration, same dependency pin manifest, same compiler/interpreter version declaration, and same build flags produce a byte-identical output artifact. The receipt is the proof. The build system is the adapter.

Software Bill of Materials (SBOM) artifacts must be receipt-bound. Dependency pin manifests must be hash-bound. Installer provenance must be checksum-verified. Build environment declarations must be canonical JSON. Any deviation from the declared environment must produce a BuildEnvironmentDriftReceipt, not a silent fallback.

Arc Reinterpretation

"A reproducible build is not a claim.

A reproducible build is a receipt.

The build system is the adapter.
The SBOM is the inventory.
The pin manifest is the contract.
The checksum is the proof.

Installer mismatch must fail closed."

Planned Releases

• v170.0 → ReproducibleBuildManifest
• v170.1 → DependencyPinManifest
• v170.2 → SBOMReceipt
• v170.3 → InstallerProvenanceReceipt
• v170.4 → BuildEnvironmentDeclarationReceipt
• v170.5 → BuildReplayEquivalenceReceipt
• v170.6 → BuildEnvironmentDriftReceipt

Expected Modules

• src/qec/analysis/reproducible_build_manifest.py
• src/qec/analysis/dependency_pin_manifest.py
• src/qec/analysis/sbom_receipts.py
• src/qec/analysis/installer_provenance_receipts.py
• src/qec/analysis/build_environment_declaration_receipts.py
• src/qec/analysis/build_replay_equivalence_receipts.py
• src/qec/analysis/build_environment_drift_receipts.py

Expected Artifacts

• ReproducibleBuildManifest
• DependencyPinManifest
• SBOMReceipt
• InstallerProvenanceReceipt
• BuildEnvironmentDeclarationReceipt
• BuildReplayEquivalenceReceipt
• BuildEnvironmentDriftReceipt

Expected Hashes

• reproducible_build_manifest_hash (v170.0)
• dependency_pin_manifest_hash (v170.1)
• sbom_receipt_hash (v170.2)
• installer_provenance_receipt_hash (v170.3)
• build_environment_declaration_receipt_hash (v170.4)
• build_replay_equivalence_receipt_hash (v170.5)
• build_environment_drift_receipt_hash (v170.6)

Core Rule

same source inputs
+ same dependency pin manifest
+ same build environment declaration
+ same compiler/interpreter version
+ same build flags
→ same output artifact bytes
→ same build_replay_equivalence_receipt_hash

Acceptance Gates

• pytest: ReproducibleBuildManifest declares source hash, dependency pin manifest hash, and build environment hash
• pytest: DependencyPinManifest rejects unpinned dependencies
• pytest: SBOMReceipt declares all direct and transitive dependencies with version and source hash
• pytest: InstallerProvenanceReceipt verifies checksum before installation; fails closed on mismatch
• pytest: BuildEnvironmentDeclarationReceipt declares OS, Python version, compiler version, and build flags
• pytest: BuildReplayEquivalenceReceipt verifies byte-identical output across declared equivalent environments
• pytest: BuildEnvironmentDriftReceipt is produced on any deviation from declared environment
• pytest: no raw curl | sh installer patterns accepted without provenance receipt
• pytest: no unpinned dependency resolution in deterministic build paths
• pytest: no live network calls in build replay tests
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: build systems are adapters, never authorities

Must Not Do

• no unpinned dependencies in deterministic build paths
• no raw installer without checksum verification
• no silent build environment fallback
• no SBOM omission
• no installer mismatch silent continuation
• no build claim without build receipt
• no version drift without drift receipt
• no hidden dependency resolution

Dependency Boundaries

• depends on: v163.x HeavyDependencyDiscoveryReceipt (dependency surface inventory)
• feeds into: v191.x Reproducible Build / Hermetic Environment Receipts v2 (hermetic extension)
• feeds into: v192.x Global Proof Composition v2 (build provenance binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v170.x implements the installer and release provenance hardening priority established in the post-v161 hygiene section. It extends the dependency governance discipline of v163.x into the build and supply-chain domain, consistent with the Reproducible Build Reminder already present in the roadmap.

---

Phase: v171.x — Deterministic Knowledge Base / Agent Memory Receipts

Status PLANNED

Source-Grounded Motivation The QEC release history contains multiple agent memory arcs: v149.1 introduced hierarchical memory arbitration, v150.0 introduced shared memory fabric, and v137.1.13 introduced memory compression and context ledger. The roadmap extension materials reference SkillOS, FAMA, and hybrid memory as research signals.

Agent memory is not a trusted store. Agent memory is a bounded, receipt-bound adapter surface. Every memory write must declare: source artifact hash, write policy, retention policy, and eviction policy. Every memory read must declare: retrieval policy, staleness policy, and confidence boundary. Every memory promotion must require a replay equivalence receipt.

The knowledge base is not an authority. It is a receipt-bound index of declared facts, each with source hash, review status, and claim scope. Facts that exceed declared evidence scope are rejected. Facts that lack source hashes are labelled UNVERIFIED.

Arc Reinterpretation

"Agent memory is not a trusted oracle.

Agent memory is a bounded receipt surface.

A memory write is a receipt.
A memory read is a bounded retrieval.
A memory promotion is a replay equivalence proof.

The knowledge base is the index.
The receipt is the boundary.
The source hash is the anchor."

Planned Releases

• v171.0 → KnowledgeBaseManifest
• v171.1 → AgentMemoryWriteReceipt
• v171.2 → AgentMemoryReadReceipt
• v171.3 → MemoryPromotionReceipt
• v171.4 → KnowledgeBaseReplayEquivalenceReceipt
• v171.5 → MemoryEvictionPolicyReceipt

Expected Modules

• src/qec/analysis/knowledge_base_manifest.py
• src/qec/analysis/agent_memory_write_receipts.py
• src/qec/analysis/agent_memory_read_receipts.py
• src/qec/analysis/memory_promotion_receipts.py
• src/qec/analysis/knowledge_base_replay_equivalence_receipts.py
• src/qec/analysis/memory_eviction_policy_receipts.py

Expected Artifacts

• KnowledgeBaseManifest
• AgentMemoryWriteReceipt
• AgentMemoryReadReceipt
• MemoryPromotionReceipt
• KnowledgeBaseReplayEquivalenceReceipt
• MemoryEvictionPolicyReceipt

Expected Hashes

• knowledge_base_manifest_hash (v171.0)
• agent_memory_write_receipt_hash (v171.1)
• agent_memory_read_receipt_hash (v171.2)
• memory_promotion_receipt_hash (v171.3)
• knowledge_base_replay_equivalence_receipt_hash (v171.4)
• memory_eviction_policy_receipt_hash (v171.5)

Core Rule

same knowledge base version
+ same source artifact hashes
+ same write policy
+ same retrieval policy
+ same retention policy
→ same canonical knowledge base state
→ same knowledge_base_replay_equivalence_receipt_hash

Acceptance Gates

• pytest: KnowledgeBaseManifest declares version, source artifact hashes, write policy, and retrieval policy
• pytest: AgentMemoryWriteReceipt declares source artifact hash, write timestamp, and retention policy
• pytest: AgentMemoryReadReceipt declares retrieval policy, staleness policy, and confidence boundary
• pytest: MemoryPromotionReceipt requires replay equivalence receipt before promotion
• pytest: KnowledgeBaseReplayEquivalenceReceipt verifies same inputs produce same knowledge base state
• pytest: MemoryEvictionPolicyReceipt declares eviction trigger, eviction scope, and rollback condition
• pytest: facts without source hashes are labelled UNVERIFIED
• pytest: facts that exceed declared evidence scope are rejected
• pytest: no live network calls in tests
• pytest: no hidden memory writes
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: memory backends are adapters, never authorities

Must Not Do

• no agent memory treated as trusted oracle
• no memory write without source artifact hash
• no memory promotion without replay equivalence receipt
• no fact without source hash
• no claim scope expansion beyond declared evidence
• no hidden memory writes
• no autonomous memory authority
• no unversioned knowledge base

Dependency Boundaries

• depends on: v165.8.x Agent Observability / Skill-Library Receipts (agent trace discipline)
• depends on: v165.6.x AI-Scientist provenance receipts (fact provenance discipline)
• feeds into: v181.x Local Agent / Tool Dispatch Receipts (memory-backed tool dispatch)
• feeds into: v187.x Human Audit / Red-Team Receipts (memory audit surface)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v171.x extends the agent observability discipline of v165.8.x into the persistent memory domain. It enforces the Agent Boundary Reminder already present in the roadmap: local dispatch only, no hidden autonomy. Memory is a bounded adapter surface, not an authority.

---

Phase: v172.x — Graphene / Photonic / Diamond / Materials Signal Receipts

Status PLANNED

Source-Grounded Motivation The roadmap extension materials reference diamond SiV sensing, Nature Photonics CV optics, QKD, and Nature Materials as research signals. The v158.x arc introduced substrate constraint contracts and material encoding receipts. The v165.9.x arc introduced quantum memory signal receipts and self-correcting memory claim boundary receipts.

v172.x extends the materials signal discipline into photonic, graphene, and diamond device contexts. These are research signal boundaries, not hardware authority claims. A graphene device signal is a source-bound claim. A photonic signal is an adapter signal. A diamond SiV sensing result is an unreviewed preprint unless declared otherwise.

The purpose of v172.x is to produce receipt-bound artifacts that declare: materials signal source, device type, measurement method, claim scope, review status, and hardware adapter boundary. No material intelligence claims. No hardware extrapolation. No physical learning language without MaterialsMemoryClaimBoundaryReceipt.

Arc Reinterpretation

"Materials may exhibit interesting dynamics.

QEC records signal boundaries only.

Physical learning language is a metaphor
unless bounded by MaterialsMemoryClaimBoundaryReceipt.

No material intelligence claims.
No hardware extrapolation.
The receipt is the boundary."

Planned Releases

• v172.0 → MaterialsSignalManifest
• v172.1 → GrapheneSignalReceipt
• v172.2 → PhotonicSignalReceipt
• v172.3 → DiamondSiVSignalReceipt
• v172.4 → MaterialsMemoryClaimBoundaryReceipt
• v172.5 → FiberOpticSignalReceipt
• v172.6 → HardwareLibraryProvenanceReceipt
• v172.7 → MaterialsReplayBoundaryReceipt

Expected Modules

• src/qec/analysis/materials_signal_manifest.py
• src/qec/analysis/graphene_signal_receipts.py
• src/qec/analysis/photonic_signal_receipts.py
• src/qec/analysis/diamond_siv_signal_receipts.py
• src/qec/analysis/materials_memory_claim_boundary_receipts.py
• src/qec/analysis/fiber_optic_signal_receipts.py
• src/qec/analysis/hardware_library_provenance_receipts.py
• src/qec/analysis/materials_replay_boundary_receipts.py

Expected Artifacts

• MaterialsSignalManifest
• GrapheneSignalReceipt
• PhotonicSignalReceipt
• DiamondSiVSignalReceipt
• MaterialsMemoryClaimBoundaryReceipt
• FiberOpticSignalReceipt
• HardwareLibraryProvenanceReceipt
• MaterialsReplayBoundaryReceipt

Expected Hashes

• materials_signal_manifest_hash (v172.0)
• graphene_signal_receipt_hash (v172.1)
• photonic_signal_receipt_hash (v172.2)
• diamond_siv_signal_receipt_hash (v172.3)
• materials_memory_claim_boundary_receipt_hash (v172.4)
• fiber_optic_signal_receipt_hash (v172.5)
• hardware_library_provenance_receipt_hash (v172.6)
• materials_replay_boundary_receipt_hash (v172.7)

Core Rule

same materials signal source
+ same declared device type
+ same declared measurement method
+ same declared claim scope
+ same review status declaration
+ same hardware adapter boundary
→ same materials_replay_boundary_receipt_hash

Acceptance Gates

• pytest: MaterialsSignalManifest declares signal source, device type, measurement method, and claim scope
• pytest: GrapheneSignalReceipt declares source paper, review status, and adapter_only=true
• pytest: PhotonicSignalReceipt declares wavelength range, loss budget, and claim scope
• pytest: DiamondSiVSignalReceipt declares spin coherence time source and UNREVIEWED_PREPRINT status if applicable
• pytest: MaterialsMemoryClaimBoundaryReceipt rejects physical learning language without explicit boundary declaration
• pytest: FiberOpticSignalReceipt declares fiber type, loss coefficient source, and adapter_only=true
• pytest: HardwareLibraryProvenanceReceipt declares library name, version, source hash, and license
• pytest: MaterialsReplayBoundaryReceipt verifies same signal source produces same canonical boundary output
• pytest: inaccessible sources are marked source_inaccessible
• pytest: no hardware authority claims
• pytest: no material intelligence claims
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: materials backends are adapters, never authorities

Must Not Do

• no material intelligence claims
• no hardware extrapolation
• no physical learning language without MaterialsMemoryClaimBoundaryReceipt
• no hardware authority claim
• no inaccessible source as sole phase justification
• no unreviewed preprint without UNREVIEWED_PREPRINT declaration
• no claim scope expansion beyond declared evidence

Dependency Boundaries

• depends on: v158.x substrate constraint contracts (material encoding discipline)
• depends on: v165.9.x quantum memory signal receipts (signal boundary discipline)
• feeds into: v185.x Photonic / Materials / Device Signal Receipts v2 (v2 extension)
• feeds into: v192.x Global Proof Composition v2 (materials signal binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v172.x extends the substrate constraint discipline of v158.x and the quantum hardware signal discipline of v165.9.x into photonic, graphene, and diamond device contexts. It enforces the Materials Signal Reminder already present in the roadmap.

---

Phase: v173.x — Interactive Proof Worlds / Citizen-Science Game Receipts

Status PLANNED

Source-Grounded Motivation The QEC release history contains multiple GameWorld arcs: v156.x introduced GameWorld intake, world adapters, observation snapshots, episode traces, strategy probes, and chaos replay verdicts. The v136.5.0 release introduced a 2D spatial arcade sandbox. The roadmap extension materials reference button-pushing explorers as a research signal.

v173.x extends the GameWorld discipline into citizen-science and interactive proof contexts. A citizen-science game is a bounded interaction surface where human participants contribute observations to a proof pipeline. The game is not the proof. The observation is the receipt. The interaction boundary is the contract.

Interactive proof worlds must declare: world adapter boundary, observation extraction policy, participant identity policy, contribution aggregation policy, and replay equivalence contract. No world authority. No participant identity leakage. No autonomous world execution.

Arc Reinterpretation

"A citizen-science game is a bounded observation surface.

Human participants contribute observations.
Observations become receipts.
Receipts enter the proof pipeline.

The game is the adapter.
The observation is the receipt.
The world is not the authority."

Planned Releases

• v173.0 → InteractiveProofWorldManifest
• v173.1 → CitizenScienceObservationReceipt
• v173.2 → ParticipantContributionBoundaryReceipt
• v173.3 → ContributionAggregationReceipt
• v173.4 → GameWorldReplayEquivalenceReceipt
• v173.5 → CitizenScienceProofReceipt

Expected Modules

• src/qec/analysis/interactive_proof_world_manifest.py
• src/qec/analysis/citizen_science_observation_receipts.py
• src/qec/analysis/participant_contribution_boundary_receipts.py
• src/qec/analysis/contribution_aggregation_receipts.py
• src/qec/analysis/game_world_replay_equivalence_receipts.py
• src/qec/analysis/citizen_science_proof_receipts.py

Expected Artifacts

• InteractiveProofWorldManifest
• CitizenScienceObservationReceipt
• ParticipantContributionBoundaryReceipt
• ContributionAggregationReceipt
• GameWorldReplayEquivalenceReceipt
• CitizenScienceProofReceipt

Expected Hashes

• interactive_proof_world_manifest_hash (v173.0)
• citizen_science_observation_receipt_hash (v173.1)
• participant_contribution_boundary_receipt_hash (v173.2)
• contribution_aggregation_receipt_hash (v173.3)
• game_world_replay_equivalence_receipt_hash (v173.4)
• citizen_science_proof_receipt_hash (v173.5)

Core Rule

same world adapter boundary
+ same observation extraction policy
+ same participant identity policy
+ same contribution aggregation policy
→ same canonical observation set
→ same game_world_replay_equivalence_receipt_hash

Acceptance Gates

• pytest: InteractiveProofWorldManifest declares world adapter boundary, observation policy, and participant identity policy
• pytest: CitizenScienceObservationReceipt is hash-stable and declares source world and extraction policy
• pytest: ParticipantContributionBoundaryReceipt declares anonymization policy and contribution scope
• pytest: ContributionAggregationReceipt declares aggregation method, ordering policy, and conflict resolution
• pytest: GameWorldReplayEquivalenceReceipt verifies same observation set produces same canonical output
• pytest: CitizenScienceProofReceipt requires aggregation receipt and replay equivalence receipt
• pytest: no participant identity leakage in receipts
• pytest: no world authority claims
• pytest: no autonomous world execution in tests
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: game worlds are adapters, never authorities

Must Not Do

• no world authority claims
• no participant identity leakage
• no autonomous world execution
• no observation accepted without extraction policy
• no aggregation without declared method
• no proof claim from game output alone
• no live world connection in tests

Dependency Boundaries

• depends on: v156.x GameWorld intake and observation contracts
• depends on: v165.8.x Agent Observability receipts (observation trace discipline)
• feeds into: v192.x Global Proof Composition v2 (citizen-science proof binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v173.x extends the GameWorld discipline of v156.x into citizen-science and interactive proof contexts. It enforces the GameWorld Reminder already present in the roadmap: observation contracts do not grant world authority.

---

Phase: v174.x — BP Dynamics / Fixed-Point Trap Receipts

Status PLANNED

Source-Grounded Motivation Belief propagation (BP) decoding dynamics appear throughout the QEC release history: v4.8.0 introduced fixed-point trap analysis, v4.9.0 introduced basin-of-attraction analysis, v5.0.0 introduced BP attractor landscape mapping, v5.1.0 introduced free-energy barrier estimation, v5.7.0 introduced BP phase-space exploration, v6.0.0 introduced spectral stability diagnostics, v6.8.0 introduced a spectral BP stability predictor, and v7.2.0 introduced a spectral instability phase map.

These earlier arcs produced diagnostic artifacts without receipt-bound invariant discipline. v174.x retrofits that discipline onto the BP dynamics surface. The purpose is to produce receipt-bound artifacts that declare: BP iteration policy, convergence criterion, fixed-point classification, attractor basin boundary, trapping set characterization, and phase-space sampling policy.

BP dynamics are not proof. BP convergence is not a correctness guarantee. Fixed-point traps are characterization artifacts, not failure modes unless declared as such. The receipt is the boundary.

Arc Reinterpretation

"BP convergence is not proof.

A fixed-point trap is a characterization artifact.
An attractor basin is a bounded region.
A spectral stability signal is an adapter signal.

The receipt declares the convergence policy.
The receipt declares the trap classification.
The receipt is the boundary."

Planned Releases

• v174.0 → BPDynamicsManifest
• v174.1 → BPIterationPolicyReceipt
• v174.2 → FixedPointTrapReceipt
• v174.3 → AttractorBasinBoundaryReceipt
• v174.4 → TrappingSetCharacterizationReceipt
• v174.5 → BPPhaseSpaceReceipt
• v174.6 → BPReplayEquivalenceReceipt

Expected Modules

• src/qec/analysis/bp_dynamics_manifest.py
• src/qec/analysis/bp_iteration_policy_receipts.py
• src/qec/analysis/fixed_point_trap_receipts.py
• src/qec/analysis/attractor_basin_boundary_receipts.py
• src/qec/analysis/trapping_set_characterization_receipts.py
• src/qec/analysis/bp_phase_space_receipts.py
• src/qec/analysis/bp_replay_equivalence_receipts.py

Expected Artifacts

• BPDynamicsManifest
• BPIterationPolicyReceipt
• FixedPointTrapReceipt
• AttractorBasinBoundaryReceipt
• TrappingSetCharacterizationReceipt
• BPPhaseSpaceReceipt
• BPReplayEquivalenceReceipt

Expected Hashes

• bp_dynamics_manifest_hash (v174.0)
• bp_iteration_policy_receipt_hash (v174.1)
• fixed_point_trap_receipt_hash (v174.2)
• attractor_basin_boundary_receipt_hash (v174.3)
• trapping_set_characterization_receipt_hash (v174.4)
• bp_phase_space_receipt_hash (v174.5)
• bp_replay_equivalence_receipt_hash (v174.6)

Core Rule

same BP iteration policy
+ same convergence criterion
+ same fixed-point classification
+ same attractor basin boundary
+ same trapping set characterization
+ same phase-space sampling policy
→ same canonical BP dynamics output
→ same bp_replay_equivalence_receipt_hash

Acceptance Gates

• pytest: BPDynamicsManifest declares iteration policy, convergence criterion, and max iteration count
• pytest: BPIterationPolicyReceipt declares damping factor, schedule, and termination condition
• pytest: FixedPointTrapReceipt classifies trap type and declares detection method
• pytest: AttractorBasinBoundaryReceipt declares basin estimation method and sampling policy
• pytest: TrappingSetCharacterizationReceipt declares trapping set type, size, and source code
• pytest: BPPhaseSpaceReceipt declares sampling grid, perturbation policy, and trajectory count
• pytest: BPReplayEquivalenceReceipt verifies same input syndrome produces same BP trajectory
• pytest: no convergence claim without declared convergence criterion
• pytest: no performance claim without benchmark receipt
• pytest: no live decoder calls in tests
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: BP backends are adapters, never authorities

Must Not Do

• no BP convergence treated as correctness proof
• no fixed-point trap treated as failure without declaration
• no attractor basin claim without estimation method
• no trapping set claim without characterization receipt
• no performance claim without benchmark receipt
• no undeclared iteration policy
• no live decoder calls in tests

Dependency Boundaries

• depends on: v166.x decoder governance receipts (canonical decoder baseline)
• depends on: v167.x qudit stabilizer receipts (qudit BP dynamics)
• feeds into: v179.x Quantum Geometry / Contextuality receipts (BP geometry signals)
• feeds into: v184.x Benchmark Ladder receipts (BP performance benchmarks)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v174.x formalizes the BP dynamics discipline established in v4.8.0 through v7.2.0. It does not replace those earlier diagnostic layers. It wraps their behavioral contracts as receipt-bound invariants, consistent with the adapter-not-authority law established in v163.x.

---

Phase: v175.x — Operator Console Unification

Status PLANNED

Source-Grounded Motivation The QEC release history contains multiple operator surface arcs: v106.x introduced a Rust TUI control surface, v107.x introduced observability dashboards and KPI health consoles, v115.x introduced phase workstation sync, v132.x introduced TUI auto-installer infrastructure, and v162.x introduced the IRC operator control surface.

These operator surfaces are fragmented across multiple arcs. v175.x unifies them under a single receipt-bound operator console manifest. The unified console is not a new execution authority. It is a bounded display and command surface that aggregates existing operator receipts, telemetry streams, and proof artifacts into a canonical operator view.

Every operator command must produce a receipt. Every display state must be derivable from canonical proof artifacts. No hidden operator side-channels. No autonomous execution authority. The console is the adapter. The receipt is the boundary.

Arc Reinterpretation

"The operator console is a bounded display surface.

It aggregates receipts.
It does not generate authority.

Every command is a receipt.
Every display state is derivable from canonical artifacts.
The console is the adapter.
The receipt is the boundary."

Planned Releases

• v175.0 → OperatorConsoleManifest
• v175.1 → ConsoleCommandReceipt
• v175.2 → ConsoleDisplayStateReceipt
• v175.3 → OperatorSessionReceipt
• v175.4 → ConsoleReplayEquivalenceReceipt

Expected Modules

• src/qec/analysis/operator_console_manifest.py
• src/qec/analysis/console_command_receipts.py
• src/qec/analysis/console_display_state_receipts.py
• src/qec/analysis/operator_session_receipts.py
• src/qec/analysis/console_replay_equivalence_receipts.py

Expected Artifacts

• OperatorConsoleManifest
• ConsoleCommandReceipt
• ConsoleDisplayStateReceipt
• OperatorSessionReceipt
• ConsoleReplayEquivalenceReceipt

Expected Hashes

• operator_console_manifest_hash (v175.0)
• console_command_receipt_hash (v175.1)
• console_display_state_receipt_hash (v175.2)
• operator_session_receipt_hash (v175.3)
• console_replay_equivalence_receipt_hash (v175.4)

Core Rule

same operator command stream
+ same canonical proof artifact set
+ same display state derivation policy
→ same canonical operator console state
→ same console_replay_equivalence_receipt_hash

Acceptance Gates

• pytest: OperatorConsoleManifest declares command surface version, display policy, and session policy
• pytest: ConsoleCommandReceipt is hash-stable and declares command type, input hash, and output hash
• pytest: ConsoleDisplayStateReceipt is derivable from canonical proof artifacts without hidden state
• pytest: OperatorSessionReceipt declares session start, end, command count, and session hash
• pytest: ConsoleReplayEquivalenceReceipt verifies same command stream produces same console state
• pytest: no hidden operator side-channels
• pytest: no autonomous execution authority
• pytest: no live network calls in tests
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: operator consoles are adapters, never authorities

Must Not Do

• no hidden operator side-channels
• no autonomous execution authority
• no display state that cannot be derived from canonical artifacts
• no operator command without receipt
• no session without session receipt
• no live network calls in tests

Dependency Boundaries

• depends on: v162.x IRC operator control surface
• depends on: v168.x proof telemetry receipts (telemetry display)
• feeds into: v190.x Operator / IRC / TUI / CLI Federation Receipts v2 (federation)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v175.x unifies the operator surface discipline established in v106.x, v107.x, v115.x, v132.x, and v162.x. It does not replace those earlier surfaces. It wraps them under a canonical operator console manifest, consistent with the Self-Hosting Reminder already present in the roadmap.

---

Phase: v176.x — Real-Time Syndrome Streaming Receipts

Status PLANNED

Source-Grounded Motivation Real-time syndrome streaming is a control-plane requirement for hardware-coupled QEC systems. The LEGO decoder architecture is cited as a research signal for real-time decoding. The roadmap extension materials reference syndrome streaming receipt schema as a Zenodo publication opportunity.

Syndrome streaming is not a live control authority. It is a bounded ingestion surface with deterministic windowing, canonical ordering, and replay-safe buffering. Every syndrome stream must declare: stream source, window policy, ordering policy, buffer policy, and replay equivalence contract. No live control authority. No safety-critical automation. No real-time guarantee without measurement receipt.

The syndrome stream manifest is the contract. The streaming window receipt is the boundary. The syndrome replay receipt is the proof.

Arc Reinterpretation

"Real-time syndrome streaming is a bounded ingestion surface.

The stream is the adapter.
The window is the boundary.
The replay receipt is the proof.

No live control authority.
No safety-critical automation.
No real-time guarantee without measurement receipt."

Planned Releases

• v176.0 → SyndromeStreamManifest
• v176.1 → StreamingWindowReceipt
• v176.2 → SyndromeReplayReceipt
• v176.3 → StreamBufferPolicyReceipt
• v176.4 → SyndromeOrderingReceipt
• v176.5 → StreamLatencyBoundaryReceipt
• v176.6 → SyndromeStreamReplayEquivalenceReceipt

Expected Modules

• src/qec/analysis/syndrome_stream_manifest.py
• src/qec/analysis/streaming_window_receipts.py
• src/qec/analysis/syndrome_replay_receipts.py
• src/qec/analysis/stream_buffer_policy_receipts.py
• src/qec/analysis/syndrome_ordering_receipts.py
• src/qec/analysis/stream_latency_boundary_receipts.py
• src/qec/analysis/syndrome_stream_replay_equivalence_receipts.py

Expected Artifacts

• SyndromeStreamManifest
• StreamingWindowReceipt
• SyndromeReplayReceipt
• StreamBufferPolicyReceipt
• SyndromeOrderingReceipt
• StreamLatencyBoundaryReceipt
• SyndromeStreamReplayEquivalenceReceipt

Expected Hashes

• syndrome_stream_manifest_hash (v176.0)
• streaming_window_receipt_hash (v176.1)
• syndrome_replay_receipt_hash (v176.2)
• stream_buffer_policy_receipt_hash (v176.3)
• syndrome_ordering_receipt_hash (v176.4)
• stream_latency_boundary_receipt_hash (v176.5)
• syndrome_stream_replay_equivalence_receipt_hash (v176.6)

Core Rule

same syndrome stream source
+ same window policy
+ same ordering policy
+ same buffer policy
+ same latency boundary declaration
→ same canonical syndrome window
→ same syndrome_stream_replay_equivalence_receipt_hash

Acceptance Gates

• pytest: SyndromeStreamManifest declares stream source, window policy, ordering policy, and buffer policy
• pytest: StreamingWindowReceipt declares window size, stride, and ordering policy
• pytest: SyndromeReplayReceipt verifies same stream produces same canonical syndrome sequence
• pytest: StreamBufferPolicyReceipt declares buffer size, overflow policy, and backpressure policy
• pytest: SyndromeOrderingReceipt declares ordering key and tie-breaking policy
• pytest: StreamLatencyBoundaryReceipt declares measured latency envelope and measurement method
• pytest: SyndromeStreamReplayEquivalenceReceipt verifies byte-identical output across replay runs
• pytest: no live hardware stream connections in tests
• pytest: no real-time guarantee without measurement receipt
• pytest: no safety-critical automation
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: stream sources are adapters, never authorities

Must Not Do

• no live control authority
• no safety-critical automation
• no real-time guarantee without measurement receipt
• no stream accepted without ordering policy
• no buffer overflow without declared overflow policy
• no latency claim without measurement receipt
• no live hardware stream in tests

Dependency Boundaries

• depends on: v166.x decoder governance receipts (syndrome input discipline)
• depends on: v174.x BP dynamics receipts (syndrome decoding dynamics)
• feeds into: v177.x Hardware Abstraction / Control-Plane Receipts (control-plane integration)
• feeds into: v192.x Global Proof Composition v2 (syndrome stream binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v176.x implements the syndrome streaming receipt schema identified as a Zenodo publication opportunity. It enforces the Syndrome Streaming Reminder already present in the roadmap: stream ingestion must be replay-safe and canonical. The qldpc_construction_receipt_hash and syndrome_stream_receipt_hash entries in the terminal proof chain are anchored here.

Phase: v177.x — Hardware Abstraction / Control-Plane Receipts

Status PLANNED

Source-Grounded Motivation Hardware abstraction is a boundary problem, not an implementation problem. QEC does not implement hardware control. QEC declares the boundary between the proof pipeline and the hardware control plane. The boundary is a receipt. The hardware is the adapter.

The QEC release history contains hardware dispatch paths in v138.4.1 (qutrit hardware dispatch), v137.1.12 (quantum route optimization), and v137.1.15 (quantum noise fidelity observatory). These earlier arcs produced hardware-adjacent artifacts without explicit control-plane boundary receipts.

v177.x formalizes the hardware abstraction boundary. Every hardware control-plane interaction must declare: hardware adapter type, control-plane protocol, latency envelope, fidelity model, noise model source, and adapter_only status. No live hardware authority. No safety-critical automation. No real-time guarantee without measurement receipt.

The control-plane manifest is the contract. The hardware adapter boundary receipt is the proof. The hardware is never the authority.

Arc Reinterpretation

"Hardware is an adapter.

The control plane is a bounded interface.
The latency envelope is a declared constraint.
The fidelity model is a source-bound signal.

No live hardware authority.
No safety-critical automation.
The receipt is the boundary."

Planned Releases

• v177.0 → HardwareAbstractionManifest
• v177.1 → ControlPlaneProtocolReceipt
• v177.2 → HardwareAdapterBoundaryReceipt
• v177.3 → LatencyEnvelopeReceipt
• v177.4 → FidelityModelBoundaryReceipt
• v177.5 → NoiseModelSourceReceipt
• v177.6 → ControlPlaneReplayEquivalenceReceipt

Expected Modules

• src/qec/analysis/hardware_abstraction_manifest.py
• src/qec/analysis/control_plane_protocol_receipts.py
• src/qec/analysis/hardware_adapter_boundary_receipts.py
• src/qec/analysis/latency_envelope_receipts.py
• src/qec/analysis/fidelity_model_boundary_receipts.py
• src/qec/analysis/noise_model_source_receipts.py
• src/qec/analysis/control_plane_replay_equivalence_receipts.py

Expected Artifacts

• HardwareAbstractionManifest
• ControlPlaneProtocolReceipt
• HardwareAdapterBoundaryReceipt
• LatencyEnvelopeReceipt
• FidelityModelBoundaryReceipt
• NoiseModelSourceReceipt
• ControlPlaneReplayEquivalenceReceipt

Expected Hashes

• hardware_abstraction_manifest_hash (v177.0)
• control_plane_protocol_receipt_hash (v177.1)
• hardware_adapter_boundary_receipt_hash (v177.2)
• latency_envelope_receipt_hash (v177.3)
• fidelity_model_boundary_receipt_hash (v177.4)
• noise_model_source_receipt_hash (v177.5)
• control_plane_replay_equivalence_receipt_hash (v177.6)

Core Rule

same hardware adapter type
+ same control-plane protocol declaration
+ same latency envelope declaration
+ same fidelity model source
+ same noise model source
→ same canonical control-plane boundary output
→ same control_plane_replay_equivalence_receipt_hash

Acceptance Gates

• pytest: HardwareAbstractionManifest declares hardware adapter type, protocol, and adapter_only=true
• pytest: ControlPlaneProtocolReceipt declares protocol version, message format, and ordering policy
• pytest: HardwareAdapterBoundaryReceipt has adapter_only=true and declares boundary type
• pytest: LatencyEnvelopeReceipt declares measured latency distribution and measurement method
• pytest: FidelityModelBoundaryReceipt declares fidelity model source, version, and claim scope
• pytest: NoiseModelSourceReceipt declares noise model type, source paper, and review status
• pytest: ControlPlaneReplayEquivalenceReceipt verifies same control-plane input produces same boundary output
• pytest: no live hardware connections in tests
• pytest: no safety-critical automation
• pytest: no real-time guarantee without measurement receipt
• pytest: no hardware authority claims
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: hardware backends are adapters, never authorities

Must Not Do

• no live hardware authority
• no safety-critical automation
• no real-time guarantee without measurement receipt
• no hardware authority claim
• no fidelity model without source declaration
• no noise model without source declaration
• no latency claim without measurement receipt
• no control-plane interaction without receipt

Dependency Boundaries

• depends on: v176.x syndrome streaming receipts (syndrome delivery to control plane)
• depends on: v165.9.x quantum hardware signal receipts (hardware signal discipline)
• feeds into: v178.x Fault-Tolerant Resource Accounting Receipts (resource overhead)
• feeds into: v189.x Cross-Environment Hardware/OS Replay Receipts (hardware environment replay)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v177.x implements the control_plane_manifest_hash entry in the terminal proof chain. It enforces the Hardware Profile Reminder already present in the roadmap: hardware profiles are adapter contexts, not authorities. The Real-Time Control Reminder is also enforced: no live control authority, no safety-critical automation.

---

Phase: v178.x — Fault-Tolerant Resource Accounting Receipts

Status PLANNED

Source-Grounded Motivation Fault-tolerant quantum computation requires explicit resource accounting: magic state distillation overhead, logical cycle overhead, physical qubit count, gate synthesis overhead, and error budget allocation. These quantities appear in the QEC release history in v178.0 (DistillationOverheadReceipt), v178.1 (LogicalCycleOverheadReceipt), and v178.2 (ResourceBudgetReplayReceipt).

Resource accounting is not a performance claim. It is a bounded receipt surface. Every overhead claim must declare: resource model source, accounting method, error budget allocation, distillation protocol, and benchmark corpus. No overhead claim without accounting receipt. No resource budget without replay equivalence proof.

The resource budget is the contract. The accounting receipt is the proof. The overhead is not the authority.

Arc Reinterpretation

"Resource overhead is not a performance claim.

Resource overhead is a bounded accounting receipt.

The resource model is the adapter.
The accounting method is the contract.
The receipt is the boundary.

No overhead claim without accounting receipt.
No resource budget without replay equivalence proof."

Planned Releases

• v178.0 → DistillationOverheadReceipt
• v178.1 → LogicalCycleOverheadReceipt
• v178.2 → ResourceBudgetReplayReceipt
• v178.3 → PhysicalQubitCountReceipt
• v178.4 → GateSynthesisOverheadReceipt
• v178.5 → ErrorBudgetAllocationReceipt
• v178.6 → ResourceAccountingManifest
• v178.7 → ResourceReplayEquivalenceReceipt

Expected Modules

• src/qec/analysis/distillation_overhead_receipts.py
• src/qec/analysis/logical_cycle_overhead_receipts.py
• src/qec/analysis/resource_budget_replay_receipts.py
• src/qec/analysis/physical_qubit_count_receipts.py
• src/qec/analysis/gate_synthesis_overhead_receipts.py
• src/qec/analysis/error_budget_allocation_receipts.py
• src/qec/analysis/resource_accounting_manifest.py
• src/qec/analysis/resource_replay_equivalence_receipts.py

Expected Artifacts

• DistillationOverheadReceipt
• LogicalCycleOverheadReceipt
• ResourceBudgetReplayReceipt
• PhysicalQubitCountReceipt
• GateSynthesisOverheadReceipt
• ErrorBudgetAllocationReceipt
• ResourceAccountingManifest
• ResourceReplayEquivalenceReceipt

Expected Hashes

• distillation_overhead_receipt_hash (v178.0)
• logical_cycle_overhead_receipt_hash (v178.1)
• resource_budget_replay_receipt_hash (v178.2)
• physical_qubit_count_receipt_hash (v178.3)
• gate_synthesis_overhead_receipt_hash (v178.4)
• error_budget_allocation_receipt_hash (v178.5)
• resource_accounting_manifest_hash (v178.6)
• resource_replay_equivalence_receipt_hash (v178.7)

Core Rule

same resource model source
+ same accounting method
+ same error budget allocation
+ same distillation protocol declaration
+ same benchmark corpus
→ same canonical resource accounting output
→ same resource_replay_equivalence_receipt_hash

Acceptance Gates

• pytest: DistillationOverheadReceipt declares distillation protocol, magic state type, and overhead factor source
• pytest: LogicalCycleOverheadReceipt declares code distance, cycle count, and logical error rate source
• pytest: ResourceBudgetReplayReceipt verifies same resource model produces same budget output
• pytest: PhysicalQubitCountReceipt declares qubit layout, connectivity model, and count method
• pytest: GateSynthesisOverheadReceipt declares synthesis algorithm, target gate set, and overhead source
• pytest: ErrorBudgetAllocationReceipt declares budget components, allocation method, and total budget
• pytest: ResourceAccountingManifest declares all component receipts and their hash chain
• pytest: ResourceReplayEquivalenceReceipt verifies byte-identical output across replay runs
• pytest: no overhead claim without accounting receipt
• pytest: no resource budget without replay equivalence proof
• pytest: no hardware authority claims
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: resource models are adapters, never authorities

Must Not Do

• no overhead claim without accounting receipt
• no resource budget without replay equivalence proof
• no hardware authority claim
• no distillation protocol without source declaration
• no error budget without allocation method
• no performance marketing without benchmark receipt
• no silent resource model assumption

Dependency Boundaries

• depends on: v177.x hardware abstraction receipts (hardware resource context)
• depends on: v166.x decoder governance receipts (logical cycle context)
• feeds into: v178.5.x IEEE 754 precision receipts (float-bearing resource metrics)
• feeds into: v184.x Benchmark Ladder receipts (resource overhead benchmarks)
• feeds into: v192.x Global Proof Composition v2 (resource accounting binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v178.x implements the resource_overhead_receipt_hash entry in the terminal proof chain. It enforces the Resource Accounting Reminder already present in the roadmap: overhead claims require accounting receipts. The three seed releases (v178.0, v178.1, v178.2) already appear in the roadmap; this section completes them with the full module, artifact, hash, and acceptance gate structure.

---

Phase: v179.x — Quantum Geometry / Contextuality / Topological Toolkit Receipts

Status PLANNED

Source-Grounded Motivation Contextuality and topological quantum codes are active research areas. The roadmap extension materials reference contextuality/code-switching and quantum geometry toolkit as research signals. The v137.8.0 arc introduced a topological graph kernel. The v137.1.16 arc introduced a geometry and topology reasoning layer. The v165.9.x arc introduced quantum geometry signal receipts.

Contextuality is a threshold phenomenon, not an ontological claim. A contextuality classification is a thresholded receipt, not a proof of quantum advantage. Topological boundaries are declared constraints, not physical truths. Geometry signals are source-bound adapter signals.

v179.x extends the three seed releases (v179.0 ContextualityThresholdReceipt, v179.1 TopologicalBoundaryReceipt, v179.2 GeometryReplayValidationReceipt) with full module, artifact, hash, and acceptance gate structure.

Arc Reinterpretation

"Contextuality is a threshold classification.

It is not an ontological claim.
It is not a quantum advantage proof.
It is a receipt with a declared sigma threshold.

Topological boundaries are declared constraints.
Geometry signals are adapter signals.
The receipt is the boundary."

Planned Releases

• v179.0 → ContextualityThresholdReceipt
• v179.1 → TopologicalBoundaryReceipt
• v179.2 → GeometryReplayValidationReceipt
• v179.3 → CodeSwitchingBoundaryReceipt
• v179.4 → TopologicalCodeManifest
• v179.5 → ContextualityReplayEquivalenceReceipt

Expected Modules

• src/qec/analysis/contextuality_threshold_receipts.py
• src/qec/analysis/topological_boundary_receipts.py
• src/qec/analysis/geometry_replay_validation_receipts.py
• src/qec/analysis/code_switching_boundary_receipts.py
• src/qec/analysis/topological_code_manifest.py
• src/qec/analysis/contextuality_replay_equivalence_receipts.py

Expected Artifacts

• ContextualityThresholdReceipt
• TopologicalBoundaryReceipt
• GeometryReplayValidationReceipt
• CodeSwitchingBoundaryReceipt
• TopologicalCodeManifest
• ContextualityReplayEquivalenceReceipt

Expected Hashes

• contextuality_threshold_receipt_hash (v179.0)
• topological_boundary_receipt_hash (v179.1)
• geometry_replay_validation_receipt_hash (v179.2)
• code_switching_boundary_receipt_hash (v179.3)
• topological_code_manifest_hash (v179.4)
• contextuality_replay_equivalence_receipt_hash (v179.5)

Core Rule

same contextuality classification method
+ same sigma threshold declaration
+ same topological boundary declaration
+ same geometry signal source
+ same code-switching policy
→ same canonical contextuality output
→ same contextuality_replay_equivalence_receipt_hash

Acceptance Gates

• pytest: ContextualityThresholdReceipt declares classification method, sigma threshold, and evidence scope
• pytest: TopologicalBoundaryReceipt declares boundary type, code family, and constraint source
• pytest: GeometryReplayValidationReceipt verifies same geometry signal produces same canonical output
• pytest: CodeSwitchingBoundaryReceipt declares switching protocol, boundary conditions, and overhead source
• pytest: TopologicalCodeManifest declares code family, parameters, and geometry model
• pytest: ContextualityReplayEquivalenceReceipt verifies byte-identical output across replay runs
• pytest: no contextuality claim without sigma threshold declaration
• pytest: no quantum advantage claim without source-bound evidence
• pytest: no topological boundary without declared constraint source
• pytest: no code-switching without boundary receipt
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: geometry backends are adapters, never authorities

Must Not Do

• no contextuality claim without sigma threshold
• no quantum advantage claim without source-bound evidence
• no ontological contextuality claim
• no topological boundary without constraint source
• no code-switching without boundary receipt
• no geometry signal without adapter_only declaration
• no cosmological geometry claim

Dependency Boundaries

• depends on: v178.5.x IEEE 754 precision receipts (float-bearing geometry metrics)
• depends on: v169.x symbolic geometry grammar (symbolic geometry signals)
• depends on: v174.x BP dynamics receipts (BP geometry signals)
• feeds into: v180.x Deterministic Quantum ML Boundary Receipts (QML geometry context)
• feeds into: v192.x Global Proof Composition v2 (contextuality binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v179.x implements the contextuality_threshold_receipt_hash entry in the terminal proof chain. It enforces the Contextuality Reminder already present in the roadmap: contextuality classifications are thresholded, not ontological. The three seed releases already appear in the roadmap; this section completes them.

---

Phase: v180.x — Deterministic Quantum ML Boundary Receipts

Status PLANNED

Source-Grounded Motivation Quantum machine learning (QML) sits at the intersection of quantum circuit execution and classical ML inference. Both domains introduce nondeterminism risks: quantum circuit sampling is inherently probabilistic, and ML inference introduces floating-point precision and quantization risks.

v180.x establishes deterministic boundary receipts for QML. A QML boundary receipt declares: circuit sampling policy, shot count, seed policy, classical post-processing policy, precision format, and equivalence contract. No QML output is accepted as proof without a boundary receipt. No QML performance claim without benchmark receipt. No quantum advantage claim without source-bound evidence.

The QML boundary is the adapter surface between quantum circuit execution and classical proof pipelines. The receipt is the boundary. The circuit is not the authority.

Arc Reinterpretation

"QML is a bounded adapter surface.

Quantum circuit outputs are probabilistic samples.
Classical post-processing introduces precision risks.
Neither is proof without a boundary receipt.

The shot count is declared.
The seed is declared.
The precision format is declared.
The receipt is the boundary."

Planned Releases

• v180.0 → QMLBoundaryManifest
• v180.1 → CircuitSamplingPolicyReceipt
• v180.2 → QMLPrecisionBoundaryReceipt
• v180.3 → QMLBenchmarkBoundaryReceipt
• v180.4 → QMLReplayEquivalenceReceipt
• v180.5 → QuantumAdvantageClaimBoundaryReceipt

Expected Modules

• src/qec/analysis/qml_boundary_manifest.py
• src/qec/analysis/circuit_sampling_policy_receipts.py
• src/qec/analysis/qml_precision_boundary_receipts.py
• src/qec/analysis/qml_benchmark_boundary_receipts.py
• src/qec/analysis/qml_replay_equivalence_receipts.py
• src/qec/analysis/quantum_advantage_claim_boundary_receipts.py

Expected Artifacts

• QMLBoundaryManifest
• CircuitSamplingPolicyReceipt
• QMLPrecisionBoundaryReceipt
• QMLBenchmarkBoundaryReceipt
• QMLReplayEquivalenceReceipt
• QuantumAdvantageClaimBoundaryReceipt

Expected Hashes

• qml_boundary_manifest_hash (v180.0)
• circuit_sampling_policy_receipt_hash (v180.1)
• qml_precision_boundary_receipt_hash (v180.2)
• qml_benchmark_boundary_receipt_hash (v180.3)
• qml_replay_equivalence_receipt_hash (v180.4)
• quantum_advantage_claim_boundary_receipt_hash (v180.5)

Core Rule

same circuit sampling policy
+ same shot count declaration
+ same seed policy
+ same classical post-processing policy
+ same precision format declaration
→ same canonical QML boundary output
→ same qml_replay_equivalence_receipt_hash

Acceptance Gates

• pytest: QMLBoundaryManifest declares circuit type, sampling policy, shot count, and seed policy
• pytest: CircuitSamplingPolicyReceipt declares shot count, seed, and sampling method
• pytest: QMLPrecisionBoundaryReceipt declares precision format and triggers ReducedPrecisionAdapterReceipt if FP16/BF16
• pytest: QMLBenchmarkBoundaryReceipt rejects any performance claim without benchmark receipt
• pytest: QMLReplayEquivalenceReceipt verifies same circuit + same seed produces same output distribution
• pytest: QuantumAdvantageClaimBoundaryReceipt requires source-bound evidence and sigma threshold
• pytest: no QML output accepted as proof without boundary receipt
• pytest: no quantum advantage claim without source-bound evidence
• pytest: no live circuit execution in tests
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: QML backends are adapters, never authorities

Must Not Do

• no QML output accepted as proof without boundary receipt
• no quantum advantage claim without source-bound evidence
• no undeclared shot count
• no undeclared seed policy
• no hidden precision format
• no live circuit execution in tests
• no hardware authority claim

Dependency Boundaries

• depends on: v178.5.x IEEE 754 precision receipts (precision format discipline)
• depends on: v179.x contextuality receipts (QML geometry context)
• depends on: v165.7.x LLM inference receipts (classical ML boundary discipline)
• feeds into: v181.x Local Agent / Tool Dispatch Receipts (QML tool dispatch)
• feeds into: v184.x Benchmark Ladder receipts (QML benchmarks)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v180.x implements the qml_boundary_receipt_hash entry in the terminal proof chain. It enforces the Contextuality Reminder and the Inference Optimization Reminder: no hardware authority claims, no quantum advantage claims without source-bound evidence.

---

Phase: v181.x — Local Agent / Tool Dispatch Receipts

Status PLANNED

Source-Grounded Motivation The QEC release history contains multiple agent dispatch arcs: v137.1.10 introduced a coding agent execution kernel, v165.8.x introduced agent observability and skill-library receipts. The roadmap extension materials reference SkillOS, FAMA, RecursiveMAS, and LLM shebang as research signals.

Local agent tool dispatch is a bounded execution surface. Every tool call must produce a receipt. Every dispatch decision must be declared before execution. No hidden tool calls. No autonomous network authority. No recursive agent execution without bounded depth declaration.

The tool dispatch surface is the adapter. The receipt is the boundary. The agent is not the authority.

v181.x is a major structural arc. It establishes the canonical tool dispatch receipt discipline that all future agent-facing phases depend on. The tool_dispatch_replay_proof_hash in the terminal proof chain is anchored here.

Arc Reinterpretation

"Local agent tool dispatch is a bounded execution surface.

Every tool call is a receipt.
Every dispatch decision is declared before execution.
Recursive depth is bounded and declared.

No hidden tool calls.
No autonomous network authority.
No recursive execution without bounded depth.

The tool is the adapter.
The receipt is the boundary.
The agent is not the authority."

Planned Releases

• v181.0 → ToolDispatchManifest
• v181.1 → ToolCallReceipt
• v181.2 → DispatchDecisionReceipt
• v181.3 → RecursiveDispatchBoundaryReceipt
• v181.4 → ToolOutputBoundaryReceipt
• v181.5 → AgentExecutionContextReceipt
• v181.6 → ToolDispatchReplayProof
• v181.7 → LocalOnlyExecutionBoundaryReceipt

Expected Modules

• src/qec/analysis/tool_dispatch_manifest.py
• src/qec/analysis/tool_call_receipts.py
• src/qec/analysis/dispatch_decision_receipts.py
• src/qec/analysis/recursive_dispatch_boundary_receipts.py
• src/qec/analysis/tool_output_boundary_receipts.py
• src/qec/analysis/agent_execution_context_receipts.py
• src/qec/analysis/tool_dispatch_replay_proof.py
• src/qec/analysis/local_only_execution_boundary_receipts.py

Expected Artifacts

• ToolDispatchManifest
• ToolCallReceipt
• DispatchDecisionReceipt
• RecursiveDispatchBoundaryReceipt
• ToolOutputBoundaryReceipt
• AgentExecutionContextReceipt
• ToolDispatchReplayProof
• LocalOnlyExecutionBoundaryReceipt

Expected Hashes

• tool_dispatch_manifest_hash (v181.0)
• tool_call_receipt_hash (v181.1)
• dispatch_decision_receipt_hash (v181.2)
• recursive_dispatch_boundary_receipt_hash (v181.3)
• tool_output_boundary_receipt_hash (v181.4)
• agent_execution_context_receipt_hash (v181.5)
• tool_dispatch_replay_proof_hash (v181.6)
• local_only_execution_boundary_receipt_hash (v181.7)

Core Rule

same tool dispatch manifest
+ same dispatch decision declaration
+ same tool call sequence
+ same recursive depth bound
+ same local-only execution policy
→ same canonical tool dispatch output
→ same tool_dispatch_replay_proof_hash

Acceptance Gates

• pytest: ToolDispatchManifest declares tool registry version, dispatch policy, and local-only flag
• pytest: ToolCallReceipt declares tool name, input hash, output hash, and execution timestamp
• pytest: DispatchDecisionReceipt declares decision type, selected tool, and selection rationale
• pytest: RecursiveDispatchBoundaryReceipt declares max recursion depth and overflow policy
• pytest: ToolOutputBoundaryReceipt declares output type, schema, and adapter_only status
• pytest: AgentExecutionContextReceipt declares agent version, context hash, and memory snapshot hash
• pytest: ToolDispatchReplayProof verifies same dispatch sequence produces same output hash
• pytest: LocalOnlyExecutionBoundaryReceipt rejects any network call outside declared local boundary
• pytest: no hidden tool calls
• pytest: no autonomous network authority
• pytest: no recursive execution without bounded depth declaration
• pytest: no tool output accepted as proof without boundary receipt
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: tools are adapters, never authorities

Must Not Do

• no hidden tool calls
• no autonomous network authority
• no recursive execution without bounded depth
• no tool output accepted as proof without boundary receipt
• no undeclared dispatch decision
• no agent execution without context receipt
• no live network calls in tests
• no unbounded recursion

Dependency Boundaries

• depends on: v165.8.x Agent Observability / Skill-Library Receipts (observability discipline)
• depends on: v171.x Knowledge Base / Agent Memory Receipts (memory-backed dispatch)
• depends on: v180.x QML Boundary Receipts (QML tool dispatch)
• feeds into: v187.x Human Audit / Red-Team Receipts (dispatch audit surface)
• feeds into: v188.x Deterministic Experiment Scheduler (scheduled dispatch)
• feeds into: v192.x Global Proof Composition v2 (dispatch proof binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v181.x implements the tool_dispatch_replay_proof_hash entry in the terminal proof chain. It is a major structural arc that enforces the Agent Boundary Reminder already present in the roadmap: local dispatch only, no hidden autonomy. It extends the agent observability discipline of v165.8.x into the full tool dispatch governance domain.

---

Phase: v182.x — Interpretability / Sparse Feature Receipts

Status PLANNED

Source-Grounded Motivation The roadmap extension materials reference Qwen-Scope SAE (Sparse Autoencoder) as a research signal for interpretability. Sparse feature decomposition of neural network activations is an active interpretability research area. QEC does not implement sparse autoencoders. QEC declares the boundary between interpretability research signals and QEC proof pipelines.

An interpretability signal is a source-bound adapter signal. A sparse feature is a characterization artifact. A feature activation is not a proof. The receipt is the boundary.

v182.x establishes receipt-bound artifacts for interpretability and sparse feature signals. Every interpretability claim must declare: model source, feature extraction method, activation basis, sparsity policy, and claim scope. No interpretability claim without boundary receipt. No feature activation treated as proof.

Arc Reinterpretation

"Interpretability is a bounded signal surface.

A sparse feature is a characterization artifact.
A feature activation is not a proof.
A model explanation is not a truth claim.

The receipt declares the extraction method.
The receipt declares the claim scope.
The receipt is the boundary."

Planned Releases

• v182.0 → InterpretabilityBoundaryManifest
• v182.1 → SparseFeatureExtractionReceipt
• v182.2 → FeatureActivationBoundaryReceipt
• v182.3 → InterpretabilityClaimScopeReceipt
• v182.4 → InterpretabilityReplayEquivalenceReceipt

Expected Modules

• src/qec/analysis/interpretability_boundary_manifest.py
• src/qec/analysis/sparse_feature_extraction_receipts.py
• src/qec/analysis/feature_activation_boundary_receipts.py
• src/qec/analysis/interpretability_claim_scope_receipts.py
• src/qec/analysis/interpretability_replay_equivalence_receipts.py

Expected Artifacts

• InterpretabilityBoundaryManifest
• SparseFeatureExtractionReceipt
• FeatureActivationBoundaryReceipt
• InterpretabilityClaimScopeReceipt
• InterpretabilityReplayEquivalenceReceipt

Expected Hashes

• interpretability_boundary_manifest_hash (v182.0)
• sparse_feature_extraction_receipt_hash (v182.1)
• feature_activation_boundary_receipt_hash (v182.2)
• interpretability_claim_scope_receipt_hash (v182.3)
• interpretability_replay_equivalence_receipt_hash (v182.4)

Core Rule

same model source
+ same feature extraction method
+ same activation basis
+ same sparsity policy
+ same claim scope declaration
→ same canonical interpretability boundary output
→ same interpretability_replay_equivalence_receipt_hash

Acceptance Gates

• pytest: InterpretabilityBoundaryManifest declares model source, extraction method, and claim scope
• pytest: SparseFeatureExtractionReceipt declares SAE architecture, sparsity target, and training corpus
• pytest: FeatureActivationBoundaryReceipt declares activation threshold, basis, and adapter_only=true
• pytest: InterpretabilityClaimScopeReceipt rejects claims that exceed declared evidence scope
• pytest: InterpretabilityReplayEquivalenceReceipt verifies same model + same input produces same feature output
• pytest: no feature activation treated as proof
• pytest: no model explanation treated as truth claim
• pytest: no live model inference in tests
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: interpretability backends are adapters, never authorities

Must Not Do

• no feature activation treated as proof
• no model explanation treated as truth claim
• no interpretability claim without boundary receipt
• no claim scope expansion beyond declared evidence
• no live model inference in tests
• no hidden extraction method
• no undeclared sparsity policy

Dependency Boundaries

• depends on: v165.7.x LLM inference receipts (model boundary discipline)
• depends on: v180.x QML boundary receipts (quantum model interpretability)
• feeds into: v187.x Human Audit / Red-Team Receipts (interpretability audit surface)
• feeds into: v192.x Global Proof Composition v2 (interpretability signal binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v182.x implements the interpretability_boundary_receipt_hash entry in the terminal proof chain. It enforces the principle that model explanations are not truth claims, consistent with the AI Scientist Reminder and the Inference Optimization Reminder already present in the roadmap.

---

Phase: v183.x — Reproducible Research Publication Receipts

Status PLANNED

Source-Grounded Motivation The roadmap extension materials identify v183.x as a Zenodo publication opportunity for reproducible publication receipts. The v165.6.x arc established AI-Scientist provenance receipts and the human review boundary discipline. The v137.17.4 arc introduced a deterministic research audit kernel.

Reproducible research publication means: every published claim is anchored to a receipt chain. The paper is not the proof. The receipt chain is the proof. Every figure must declare its data source hash. Every table must declare its computation receipt. Every claim must declare its evidence scope and review status.

v183.x is a major structural arc. It extends the provenance discipline of v165.6.x into the full publication workflow: from data collection through figure generation through claim scope declaration through human review through publication receipt.

Arc Reinterpretation

"A paper is not a proof.

A paper is a bounded claim surface
anchored to a receipt chain.

Every figure has a data source hash.
Every table has a computation receipt.
Every claim has a declared evidence scope.
Every claim has a declared review status.

Generated papers are not evidence.
The receipt chain is the proof."

Planned Releases

• v183.0 → ReproduciblePublicationManifest
• v183.1 → FigureDataSourceReceipt
• v183.2 → TableComputationReceipt
• v183.3 → ClaimEvidenceScopeReceipt
• v183.4 → PublicationHumanReviewReceipt
• v183.5 → CitationChainIntegrityReceipt
• v183.6 → PublicationReplayEquivalenceReceipt
• v183.7 → ZenodoProvenanceReceipt

Expected Modules

• src/qec/analysis/reproducible_publication_manifest.py
• src/qec/analysis/figure_data_source_receipts.py
• src/qec/analysis/table_computation_receipts.py
• src/qec/analysis/claim_evidence_scope_receipts.py
• src/qec/analysis/publication_human_review_receipts.py
• src/qec/analysis/citation_chain_integrity_receipts.py
• src/qec/analysis/publication_replay_equivalence_receipts.py
• src/qec/analysis/zenodo_provenance_receipts.py

Expected Artifacts

• ReproduciblePublicationManifest
• FigureDataSourceReceipt
• TableComputationReceipt
• ClaimEvidenceScopeReceipt
• PublicationHumanReviewReceipt
• CitationChainIntegrityReceipt
• PublicationReplayEquivalenceReceipt
• ZenodoProvenanceReceipt

Expected Hashes

• reproducible_publication_manifest_hash (v183.0)
• figure_data_source_receipt_hash (v183.1)
• table_computation_receipt_hash (v183.2)
• claim_evidence_scope_receipt_hash (v183.3)
• publication_human_review_receipt_hash (v183.4)
• citation_chain_integrity_receipt_hash (v183.5)
• publication_replay_equivalence_receipt_hash (v183.6)
• zenodo_provenance_receipt_hash (v183.7)

Core Rule

same publication artifact set
+ same figure data source hashes
+ same table computation receipts
+ same claim evidence scope declarations
+ same human review status
+ same citation chain integrity receipts
→ same reproducible_publication_receipt_hash

Acceptance Gates

• pytest: ReproduciblePublicationManifest declares all component receipts and their hash chain
• pytest: FigureDataSourceReceipt declares data source hash, processing pipeline hash, and render policy
• pytest: TableComputationReceipt declares computation receipt hash and output schema
• pytest: ClaimEvidenceScopeReceipt rejects claims that exceed declared evidence scope
• pytest: PublicationHumanReviewReceipt requires explicit reviewer identity or UNREVIEWED flag
• pytest: CitationChainIntegrityReceipt verifies each citation against a declared source list
• pytest: PublicationReplayEquivalenceReceipt verifies same inputs produce same publication artifact hashes
• pytest: ZenodoProvenanceReceipt declares DOI, upload timestamp, and artifact hash
• pytest: generated papers are rejected as evidence without human review receipt
• pytest: no citation hallucination
• pytest: no claim scope expansion beyond declared evidence
• dependency boundary: no import from src/qec/decoder/

Must Not Do

• no generated paper treated as evidence
• no citation hallucination
• no claim scope expansion beyond declared evidence
• no figure without data source receipt
• no table without computation receipt
• no claim without evidence scope declaration
• no publication without human review receipt
• no Zenodo upload without provenance receipt

Dependency Boundaries

• depends on: v165.6.x AI-Scientist provenance receipts (provenance discipline)
• depends on: v169.x symbolic geometry grammar (symbolic claim provenance)
• depends on: v171.x knowledge base receipts (fact provenance)
• feeds into: v192.x Global Proof Composition v2 (publication receipt binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v183.x implements the reproducible_publication_receipt_hash entry in the terminal proof chain. It is identified as a Zenodo publication opportunity. It enforces the AI Scientist Reminder already present in the roadmap: generated research is not evidence, no model output as source authority.

---

Phase: v184.x — Benchmark Ladder / External Comparator Receipts

Status PLANNED

Source-Grounded Motivation The roadmap extension materials reference Riverlane QEC roadmap, Google Willow, Qiskit, and Phoronix as benchmark signal sources. The risk register identifies benchmark marketing drift as an explicit risk. The v165.7.x arc established inference benchmark receipts. The v166.x arc established decoder benchmark ladder receipts.

A benchmark ladder is a bounded, receipt-bound comparison surface. Every rung of the ladder declares: comparator identity, comparator version, benchmark corpus, hardware declaration, measurement method, and claim scope. No benchmark claim without comparator receipt. No external comparator treated as authority.

v184.x is a major structural arc. It establishes the canonical benchmark ladder discipline that all performance-claiming phases depend on. The benchmark_ladder_receipt_hash in the terminal proof chain is anchored here.

Arc Reinterpretation

"A benchmark is not a proof.

A benchmark is a bounded comparison receipt.

The comparator is declared.
The corpus is declared.
The hardware is declared.
The claim scope is declared.

No benchmark claim without comparator receipt.
No external comparator treated as authority.
No benchmark marketing without replay equivalence.
The receipt is the boundary."

Planned Releases

• v184.0 → BenchmarkLadderManifest
• v184.1 → ComparatorIdentityReceipt
• v184.2 → BenchmarkCorpusReceipt
• v184.3 → HardwareBenchmarkDeclarationReceipt
• v184.4 → BenchmarkMeasurementReceipt
• v184.5 → ExternalComparatorBoundaryReceipt
• v184.6 → BenchmarkClaimScopeReceipt
• v184.7 → BenchmarkReplayEquivalenceReceipt
• v184.8 → BenchmarkLadderReceipt

Expected Modules

• src/qec/analysis/benchmark_ladder_manifest.py
• src/qec/analysis/comparator_identity_receipts.py
• src/qec/analysis/benchmark_corpus_receipts.py
• src/qec/analysis/hardware_benchmark_declaration_receipts.py
• src/qec/analysis/benchmark_measurement_receipts.py
• src/qec/analysis/external_comparator_boundary_receipts.py
• src/qec/analysis/benchmark_claim_scope_receipts.py
• src/qec/analysis/benchmark_replay_equivalence_receipts.py
• src/qec/analysis/benchmark_ladder_receipts.py

Expected Artifacts

• BenchmarkLadderManifest
• ComparatorIdentityReceipt
• BenchmarkCorpusReceipt
• HardwareBenchmarkDeclarationReceipt
• BenchmarkMeasurementReceipt
• ExternalComparatorBoundaryReceipt
• BenchmarkClaimScopeReceipt
• BenchmarkReplayEquivalenceReceipt
• BenchmarkLadderReceipt

Expected Hashes

• benchmark_ladder_manifest_hash (v184.0)
• comparator_identity_receipt_hash (v184.1)
• benchmark_corpus_receipt_hash (v184.2)
• hardware_benchmark_declaration_receipt_hash (v184.3)
• benchmark_measurement_receipt_hash (v184.4)
• external_comparator_boundary_receipt_hash (v184.5)
• benchmark_claim_scope_receipt_hash (v184.6)
• benchmark_replay_equivalence_receipt_hash (v184.7)
• benchmark_ladder_receipt_hash (v184.8)

Core Rule

same comparator identity
+ same benchmark corpus
+ same hardware declaration
+ same measurement method
+ same claim scope declaration
→ same canonical benchmark output
→ same benchmark_ladder_receipt_hash

Acceptance Gates

• pytest: BenchmarkLadderManifest declares all rungs, comparators, and their receipt hashes
• pytest: ComparatorIdentityReceipt declares comparator name, version, source, and adapter_only status
• pytest: BenchmarkCorpusReceipt declares corpus name, version, size, and source hash
• pytest: HardwareBenchmarkDeclarationReceipt declares hardware type, OS, and measurement environment
• pytest: BenchmarkMeasurementReceipt declares measurement method, repetition count, and statistical summary
• pytest: ExternalComparatorBoundaryReceipt has adapter_only=true for all external comparators
• pytest: BenchmarkClaimScopeReceipt rejects claims that exceed declared evidence scope
• pytest: BenchmarkReplayEquivalenceReceipt verifies same benchmark inputs produce same measurement output
• pytest: BenchmarkLadderReceipt aggregates all rung receipts into a canonical ladder hash
• pytest: no benchmark claim without comparator receipt
• pytest: no external comparator treated as authority
• pytest: no benchmark marketing without replay equivalence
• pytest: ParameterGolfCompressionReceipt, InferenceMemoryBandwidthReceipt, and ByteLevelModelBoundaryReceipt from v165.7.x are valid ladder inputs
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: external comparators are adapters, never authorities

Must Not Do

• no benchmark claim without comparator receipt
• no external comparator treated as authority
• no benchmark marketing without replay equivalence
• no hardware authority claim
• no claim scope expansion beyond declared evidence
• no benchmark claim without hardware declaration
• no performance marketing language
• no quantum advantage claim from benchmark alone

Dependency Boundaries

• depends on: v165.7.x inference benchmark receipts (inference benchmark inputs)
• depends on: v166.x decoder benchmark ladder receipts (decoder benchmark inputs)
• depends on: v178.x resource accounting receipts (resource overhead benchmarks)
• depends on: v180.x QML boundary receipts (QML benchmarks)
• feeds into: v192.x Global Proof Composition v2 (benchmark binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v184.x implements the benchmark_ladder_receipt_hash entry in the terminal proof chain. It enforces the Benchmark Reminder already present in the roadmap: no benchmark claims without comparator receipts. It is identified as a Zenodo publication opportunity. The three enhancement items from the existing placeholder (ParameterGolfCompressionReceipt, InferenceMemoryBandwidthReceipt, ByteLevelModelBoundaryReceipt) are incorporated as valid ladder inputs.

Phase: v185.x — Photonic / Materials / Device Signal Receipts v2

Status PLANNED

Source-Grounded Motivation v172.x established the initial materials signal receipt discipline for graphene, photonic, diamond, and fiber optic signals. v185.x is the v2 extension of that discipline, incorporating the additional device signal contexts that emerge from v177.x (hardware abstraction), v178.x (resource accounting), and v179.x (contextuality/topological toolkit).

The v2 extension adds: device-level signal receipts for integrated photonic circuits, continuous-variable (CV) optics signals (from Nature Photonics CV optics research signal), quantum key distribution (QKD) signal receipts, and hardware library provenance receipts for device-specific software stacks. All signals remain source-bound adapter signals. No device authority. No hardware extrapolation.

The materials_signal_boundary_receipt_v2_hash in the terminal proof chain is anchored here.

Arc Reinterpretation

"Device signals are adapter signals.

Integrated photonic circuits are adapters.
CV optics signals are source-bound claims.
QKD signals are source-bound claims.
Hardware library stacks are adapters.

No device authority.
No hardware extrapolation.
The receipt is the boundary."

Planned Releases

• v185.0 → MaterialsSignalManifestV2
• v185.1 → IntegratedPhotonicCircuitReceipt
• v185.2 → CVOpticsSignalReceipt
• v185.3 → QKDSignalReceipt
• v185.4 → DeviceStackProvenanceReceipt
• v185.5 → MaterialsMemoryClaimBoundaryReceiptV2
• v185.6 → MaterialsSignalBoundaryReceiptV2

Expected Modules

• src/qec/analysis/materials_signal_manifest_v2.py
• src/qec/analysis/integrated_photonic_circuit_receipts.py
• src/qec/analysis/cv_optics_signal_receipts.py
• src/qec/analysis/qkd_signal_receipts.py
• src/qec/analysis/device_stack_provenance_receipts.py
• src/qec/analysis/materials_memory_claim_boundary_receipts_v2.py
• src/qec/analysis/materials_signal_boundary_receipts_v2.py

Expected Artifacts

• MaterialsSignalManifestV2
• IntegratedPhotonicCircuitReceipt
• CVOpticsSignalReceipt
• QKDSignalReceipt
• DeviceStackProvenanceReceipt
• MaterialsMemoryClaimBoundaryReceiptV2
• MaterialsSignalBoundaryReceiptV2

Expected Hashes

• materials_signal_manifest_v2_hash (v185.0)
• integrated_photonic_circuit_receipt_hash (v185.1)
• cv_optics_signal_receipt_hash (v185.2)
• qkd_signal_receipt_hash (v185.3)
• device_stack_provenance_receipt_hash (v185.4)
• materials_memory_claim_boundary_receipt_v2_hash (v185.5)
• materials_signal_boundary_receipt_v2_hash (v185.6)

Core Rule

same device signal source
+ same declared device type
+ same declared measurement method
+ same declared claim scope
+ same hardware adapter boundary
+ same device stack provenance
→ same materials_signal_boundary_receipt_v2_hash

Acceptance Gates

• pytest: MaterialsSignalManifestV2 declares all component signal receipts and their hash chain
• pytest: IntegratedPhotonicCircuitReceipt declares circuit topology, loss model, and adapter_only=true
• pytest: CVOpticsSignalReceipt declares squeezing parameter source, measurement basis, and claim scope
• pytest: QKDSignalReceipt declares protocol version, key rate source, and UNREVIEWED_PREPRINT status if applicable
• pytest: DeviceStackProvenanceReceipt declares library name, version, source hash, and license
• pytest: MaterialsMemoryClaimBoundaryReceiptV2 rejects physical learning language without explicit boundary
• pytest: MaterialsSignalBoundaryReceiptV2 aggregates all v2 signal receipts into canonical boundary hash
• pytest: inaccessible sources are marked source_inaccessible
• pytest: no hardware authority claims
• pytest: no device authority claims
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: device backends are adapters, never authorities

Must Not Do

• no device authority claims
• no hardware extrapolation
• no physical learning language without boundary receipt
• no QKD key rate claim without source declaration
• no CV optics claim without measurement basis declaration
• no inaccessible source as sole phase justification
• no hardware authority claim

Dependency Boundaries

• depends on: v172.x materials signal receipts (v1 discipline)
• depends on: v177.x hardware abstraction receipts (hardware adapter boundary)
• depends on: v178.x resource accounting receipts (device resource context)
• feeds into: v192.x Global Proof Composition v2 (materials signal v2 binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v185.x implements the materials_signal_boundary_receipt_v2_hash entry in the terminal proof chain. It extends the v172.x materials signal discipline with the device-level signal contexts that emerge from the v177.x–v179.x arc cluster. The three enhancement items from the existing placeholder (MaterialsMemoryClaimBoundaryReceipt, FiberOpticSignalReceipt, HardwareLibraryProvenanceReceipt) are incorporated as v2 artifacts.

---

Phase: v186.x — Symbolic Diagram Compiler v2 / Grammar Extension

Status PLANNED

Source-Grounded Motivation The QEC repository contains qec_theory_diagram.txt as a symbolic grammar reference. v169.x established the symbolic geometry grammar and cosmovirus sandbox discipline. v186.x is the v2 extension of the symbolic diagram compiler, incorporating the grammar extensions that emerge from v169.x (symbolic hypothesis receipts), v179.x (topological code manifests), and v183.x (reproducible publication receipts).

The symbolic diagram compiler v2 extends the tokenizer, compiler manifest, and boundary replay receipts with: topological code diagram support, reproducible figure generation, and grammar extension receipts. The compiler is not an authority. The compiled diagram is not a proof. The receipt is the boundary.

The symbolic_compiler_manifest_hash in the terminal proof chain is anchored here.

Arc Reinterpretation

"The symbolic diagram compiler is a bounded rendering surface.

A compiled diagram is not a proof.
A grammar extension is not a theorem.
A topological code diagram is a visualization artifact.

claim_mode = SYMBOLIC_ONLY is mandatory.
The receipt is the boundary."

Planned Releases

• v186.0 → SymbolicTokenizerReceiptV2
• v186.1 → DiagramCompilerManifestV2
• v186.2 → SymbolicBoundaryReplayReceiptV2
• v186.3 → TopologicalDiagramReceipt
• v186.4 → GrammarExtensionReceipt
• v186.5 → ReproducibleFigureReceipt
• v186.6 → SymbolicCompilerManifest

Expected Modules

• src/qec/analysis/symbolic_tokenizer_receipts_v2.py
• src/qec/analysis/diagram_compiler_manifest_v2.py
• src/qec/analysis/symbolic_boundary_replay_receipts_v2.py
• src/qec/analysis/topological_diagram_receipts.py
• src/qec/analysis/grammar_extension_receipts.py
• src/qec/analysis/reproducible_figure_receipts.py
• src/qec/analysis/symbolic_compiler_manifest.py

Expected Artifacts

• SymbolicTokenizerReceiptV2
• DiagramCompilerManifestV2
• SymbolicBoundaryReplayReceiptV2
• TopologicalDiagramReceipt
• GrammarExtensionReceipt
• ReproducibleFigureReceipt
• SymbolicCompilerManifest

Expected Hashes

• symbolic_tokenizer_receipt_v2_hash (v186.0)
• diagram_compiler_manifest_v2_hash (v186.1)
• symbolic_boundary_replay_receipt_v2_hash (v186.2)
• topological_diagram_receipt_hash (v186.3)
• grammar_extension_receipt_hash (v186.4)
• reproducible_figure_receipt_hash (v186.5)
• symbolic_compiler_manifest_hash (v186.6)

Core Rule

same grammar version
+ same tokenizer policy
+ same compiler manifest
+ same claim_mode = SYMBOLIC_ONLY
+ same input symbolic expression
→ same compiled diagram output
→ same symbolic_compiler_manifest_hash

Acceptance Gates

• pytest: SymbolicTokenizerReceiptV2 declares grammar version, token policy, and encoding
• pytest: DiagramCompilerManifestV2 declares compiler version, grammar extensions, and output format
• pytest: SymbolicBoundaryReplayReceiptV2 verifies same symbolic input produces same diagram output
• pytest: TopologicalDiagramReceipt declares code family, diagram type, and claim_mode = SYMBOLIC_ONLY
• pytest: GrammarExtensionReceipt declares extension name, grammar rule, and backward compatibility status
• pytest: ReproducibleFigureReceipt declares data source hash, render policy, and output format
• pytest: SymbolicCompilerManifest aggregates all v2 compiler receipts into canonical manifest hash
• pytest: claim_mode = SYMBOLIC_ONLY is enforced for all symbolic surfaces
• pytest: no diagram treated as proof
• pytest: no grammar extension treated as theorem
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: diagram renderers are adapters, never authorities

Must Not Do

• no diagram treated as proof
• no grammar extension treated as theorem
• no claim_mode omission
• no symbolic expression without claim_mode declaration
• no figure without data source receipt
• no grammar extension without backward compatibility declaration
• no compiler output without boundary receipt

Dependency Boundaries

• depends on: v169.x symbolic geometry grammar (v1 discipline)
• depends on: v179.x topological code manifests (topological diagram inputs)
• depends on: v183.x reproducible publication receipts (figure generation discipline)
• feeds into: v192.x Global Proof Composition v2 (symbolic compiler binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v186.x implements the symbolic_compiler_manifest_hash entry in the terminal proof chain. It is identified as a Zenodo publication opportunity. The three seed releases (v186.0, v186.1, v186.2) already appear in the roadmap; this section completes them with the full module, artifact, hash, and acceptance gate structure.

---

Phase: v187.x — Human Audit / Red-Team Receipts

Status PLANNED

Source-Grounded Motivation Human audit and red-teaming are bounded review surfaces. Every audit finding must be receipt-bound. Every red-team exercise must declare: scope, methodology, reviewer identity policy, finding classification, and remediation receipt. No audit finding treated as proof of correctness. No red-team exercise treated as security guarantee.

The QEC release history contains audit layers in v137.10.3 (claim audit kernel), v137.17.4 (deterministic research audit kernel), v137.17.6 (deterministic ledger replay certification), and v137.18.2 (governance memory / I-O boundary auditor). These earlier arcs produced audit artifacts without explicit human review boundary receipts.

v187.x formalizes the human audit and red-team receipt discipline. The audit_trail_receipt_hash in the terminal proof chain is anchored here.

Arc Reinterpretation

"Human audit is a bounded review surface.

An audit finding is a receipt.
A red-team result is a bounded claim.
A remediation is a declared contract.

No audit finding treated as proof of correctness.
No red-team exercise treated as security guarantee.
The receipt is the boundary."

Planned Releases

• v187.0 → HumanAuditManifest
• v187.1 → AuditFindingReceipt
• v187.2 → RedTeamScopeReceipt
• v187.3 → RedTeamFindingReceipt
• v187.4 → RemediationContractReceipt
• v187.5 → AuditTrailReceipt
• v187.6 → AuditReplayEquivalenceReceipt

Expected Modules

• src/qec/analysis/human_audit_manifest.py
• src/qec/analysis/audit_finding_receipts.py
• src/qec/analysis/red_team_scope_receipts.py
• src/qec/analysis/red_team_finding_receipts.py
• src/qec/analysis/remediation_contract_receipts.py
• src/qec/analysis/audit_trail_receipts.py
• src/qec/analysis/audit_replay_equivalence_receipts.py

Expected Artifacts

• HumanAuditManifest
• AuditFindingReceipt
• RedTeamScopeReceipt
• RedTeamFindingReceipt
• RemediationContractReceipt
• AuditTrailReceipt
• AuditReplayEquivalenceReceipt

Expected Hashes

• human_audit_manifest_hash (v187.0)
• audit_finding_receipt_hash (v187.1)
• red_team_scope_receipt_hash (v187.2)
• red_team_finding_receipt_hash (v187.3)
• remediation_contract_receipt_hash (v187.4)
• audit_trail_receipt_hash (v187.5)
• audit_replay_equivalence_receipt_hash (v187.6)

Core Rule

same audit scope declaration
+ same reviewer identity policy
+ same finding classification policy
+ same remediation contract policy
→ same canonical audit trail output
→ same audit_trail_receipt_hash

Acceptance Gates

• pytest: HumanAuditManifest declares audit scope, reviewer identity policy, and finding classification
• pytest: AuditFindingReceipt declares finding type, severity, evidence hash, and reviewer identity
• pytest: RedTeamScopeReceipt declares scope boundaries, methodology, and out-of-scope declarations
• pytest: RedTeamFindingReceipt declares finding type, reproduction steps hash, and severity
• pytest: RemediationContractReceipt declares remediation action, deadline, and verification method
• pytest: AuditTrailReceipt aggregates all finding receipts into canonical audit trail hash
• pytest: AuditReplayEquivalenceReceipt verifies same audit inputs produce same finding set
• pytest: no audit finding treated as proof of correctness
• pytest: no red-team result treated as security guarantee
• pytest: reviewer identity is declared or ANONYMOUS flag is set
• dependency boundary: no import from src/qec/decoder/

Must Not Do

• no audit finding treated as proof of correctness
• no red-team result treated as security guarantee
• no undeclared audit scope
• no finding without evidence hash
• no remediation without contract receipt
• no audit trail without aggregation receipt
• no reviewer identity leakage beyond declared policy

Dependency Boundaries

• depends on: v165.6.x AI-Scientist provenance receipts (claim audit discipline)
• depends on: v171.x knowledge base receipts (fact audit surface)
• depends on: v181.x tool dispatch receipts (dispatch audit surface)
• depends on: v182.x interpretability receipts (interpretability audit surface)
• feeds into: v192.x Global Proof Composition v2 (audit trail binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v187.x implements the audit_trail_receipt_hash entry in the terminal proof chain. It enforces the human review boundary discipline established in v165.6.x and extends it into the full audit and red-team domain.

---

Phase: v188.x — Deterministic Experiment Scheduler

Status PLANNED

Source-Grounded Motivation Deterministic experiment scheduling is a bounded execution surface. Every scheduled experiment must declare: experiment identity, input corpus hash, dependency receipts, execution policy, feedback loop stability policy, and replay equivalence contract. No live control authority. No safety-critical automation. No real-time guarantee without measurement receipt.

The roadmap extension materials reference FeedbackLoopStabilityReceipt, RealTimeBoundaryReceipt, TransportBackpressureReceipt, and CollapsePreventionSignalReceipt as enhancement items. These are incorporated as full releases.

The experiment_scheduler_replay_proof_hash in the terminal proof chain is anchored here.

Arc Reinterpretation

"The experiment scheduler is a bounded execution surface.

Every scheduled experiment is a receipt.
Every feedback loop is bounded.
Every transport backpressure event is declared.

No live control authority.
No safety-critical automation.
No real-time guarantee without measurement receipt.
The receipt is the boundary."

Planned Releases

• v188.0 → ExperimentSchedulerManifest
• v188.1 → ScheduledExperimentReceipt
• v188.2 → FeedbackLoopStabilityReceipt
• v188.3 → RealTimeBoundaryReceipt
• v188.4 → TransportBackpressureReceipt
• v188.5 → CollapsePreventionSignalReceipt
• v188.6 → ExperimentDependencyReceipt
• v188.7 → ExperimentSchedulerReplayProof

Expected Modules

• src/qec/analysis/experiment_scheduler_manifest.py
• src/qec/analysis/scheduled_experiment_receipts.py
• src/qec/analysis/feedback_loop_stability_receipts.py
• src/qec/analysis/real_time_boundary_receipts.py
• src/qec/analysis/transport_backpressure_receipts.py
• src/qec/analysis/collapse_prevention_signal_receipts.py
• src/qec/analysis/experiment_dependency_receipts.py
• src/qec/analysis/experiment_scheduler_replay_proof.py

Expected Artifacts

• ExperimentSchedulerManifest
• ScheduledExperimentReceipt
• FeedbackLoopStabilityReceipt
• RealTimeBoundaryReceipt
• TransportBackpressureReceipt
• CollapsePreventionSignalReceipt
• ExperimentDependencyReceipt
• ExperimentSchedulerReplayProof

Expected Hashes

• experiment_scheduler_manifest_hash (v188.0)
• scheduled_experiment_receipt_hash (v188.1)
• feedback_loop_stability_receipt_hash (v188.2)
• real_time_boundary_receipt_hash (v188.3)
• transport_backpressure_receipt_hash (v188.4)
• collapse_prevention_signal_receipt_hash (v188.5)
• experiment_dependency_receipt_hash (v188.6)
• experiment_scheduler_replay_proof_hash (v188.7)

Core Rule

same experiment scheduler manifest
+ same scheduled experiment sequence
+ same feedback loop stability policy
+ same transport backpressure policy
+ same dependency receipts
→ same canonical experiment schedule output
→ same experiment_scheduler_replay_proof_hash

Acceptance Gates

• pytest: ExperimentSchedulerManifest declares scheduler version, execution policy, and local-only flag
• pytest: ScheduledExperimentReceipt declares experiment identity, input corpus hash, and dependency receipts
• pytest: FeedbackLoopStabilityReceipt declares loop bound, stability criterion, and overflow policy
• pytest: RealTimeBoundaryReceipt declares latency envelope, measurement method, and adapter_only=true
• pytest: TransportBackpressureReceipt declares backpressure trigger, buffer policy, and recovery policy
• pytest: CollapsePreventionSignalReceipt declares signal type, threshold, and intervention policy
• pytest: ExperimentDependencyReceipt declares all upstream receipts required before execution
• pytest: ExperimentSchedulerReplayProof verifies same schedule produces same execution receipt sequence
• pytest: no live control authority
• pytest: no safety-critical automation
• pytest: no real-time guarantee without measurement receipt
• pytest: no unbounded feedback loops
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: schedulers are adapters, never authorities

Must Not Do

• no live control authority
• no safety-critical automation
• no real-time guarantee without measurement receipt
• no unbounded feedback loops
• no experiment without dependency receipts
• no transport backpressure without declared policy
• no collapse prevention without signal receipt
• no live network calls in tests

Dependency Boundaries

• depends on: v181.x tool dispatch receipts (scheduled tool dispatch)
• depends on: v176.x syndrome streaming receipts (syndrome-driven scheduling)
• depends on: v183.x reproducible publication receipts (experiment publication workflow)
• feeds into: v189.x Cross-Environment Hardware/OS Replay Receipts (scheduled cross-environment replay)
• feeds into: v192.x Global Proof Composition v2 (experiment scheduler binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v188.x implements the experiment_scheduler_replay_proof_hash entry in the terminal proof chain. It enforces the Real-Time Control Reminder already present in the roadmap. The four enhancement items from the existing placeholder (FeedbackLoopStabilityReceipt, RealTimeBoundaryReceipt, TransportBackpressureReceipt, CollapsePreventionSignalReceipt) are incorporated as full releases.

---

Phase: v189.x — Cross-Environment Hardware/OS Replay Receipts

Status PLANNED

Source-Grounded Motivation Cross-environment replay is a foundational QEC requirement. The v148.6 arc introduced a cross-environment replay kernel. The v151.8 arc introduced replay and cross-environment resonance proof. These earlier arcs established the core replay equivalence discipline.

v189.x is a major structural arc. It extends the cross-environment replay discipline into hardware and OS contexts: different CPU architectures, different operating systems, different Python versions, different NumPy versions, different BLAS implementations, and different floating-point hardware behaviors. Every cross-environment replay must declare: environment manifest, hardware declaration, OS declaration, dependency version manifest, and equivalence contract.

The cross_environment_replay_receipt_hash in the terminal proof chain is anchored here.

Arc Reinterpretation

"Cross-environment replay is not a claim.

Cross-environment replay is a proof.

The environment manifest is the contract.
The hardware declaration is the adapter boundary.
The OS declaration is the execution context.
The equivalence contract is the proof.

same inputs
+ same declared environment
→ same outputs
→ same cross_environment_replay_receipt_hash

Deviation from declared environment
→ EnvironmentDriftReceipt
→ not silent fallback."

Planned Releases

• v189.0 → CrossEnvironmentReplayManifest
• v189.1 → EnvironmentDeclarationReceipt
• v189.2 → HardwareEnvironmentBoundaryReceipt
• v189.3 → OSEnvironmentBoundaryReceipt
• v189.4 → DependencyVersionEquivalenceReceipt
• v189.5 → FloatingPointEnvironmentReceipt
• v189.6 → EnvironmentDriftReceipt
• v189.7 → CrossEnvironmentReplayReceipt
• v189.8 → CrossEnvironmentReplayProof

Expected Modules

• src/qec/analysis/cross_environment_replay_manifest.py
• src/qec/analysis/environment_declaration_receipts.py
• src/qec/analysis/hardware_environment_boundary_receipts.py
• src/qec/analysis/os_environment_boundary_receipts.py
• src/qec/analysis/dependency_version_equivalence_receipts.py
• src/qec/analysis/floating_point_environment_receipts.py
• src/qec/analysis/environment_drift_receipts.py
• src/qec/analysis/cross_environment_replay_receipts.py
• src/qec/analysis/cross_environment_replay_proof.py

Expected Artifacts

• CrossEnvironmentReplayManifest
• EnvironmentDeclarationReceipt
• HardwareEnvironmentBoundaryReceipt
• OSEnvironmentBoundaryReceipt
• DependencyVersionEquivalenceReceipt
• FloatingPointEnvironmentReceipt
• EnvironmentDriftReceipt
• CrossEnvironmentReplayReceipt
• CrossEnvironmentReplayProof

Expected Hashes

• cross_environment_replay_manifest_hash (v189.0)
• environment_declaration_receipt_hash (v189.1)
• hardware_environment_boundary_receipt_hash (v189.2)
• os_environment_boundary_receipt_hash (v189.3)
• dependency_version_equivalence_receipt_hash (v189.4)
• floating_point_environment_receipt_hash (v189.5)
• environment_drift_receipt_hash (v189.6)
• cross_environment_replay_receipt_hash (v189.7)
• cross_environment_replay_proof_hash (v189.8)

Core Rule

same canonical inputs
+ same declared environment manifest
+ same hardware declaration
+ same OS declaration
+ same dependency version manifest
+ same floating-point environment declaration
→ same canonical outputs
→ same cross_environment_replay_receipt_hash

Any deviation from declared environment
→ EnvironmentDriftReceipt produced
→ not silent fallback

Acceptance Gates

• pytest: CrossEnvironmentReplayManifest declares all environment components and their receipt hashes
• pytest: EnvironmentDeclarationReceipt declares CPU architecture, OS, Python version, and BLAS implementation
• pytest: HardwareEnvironmentBoundaryReceipt declares hardware type, FPU behavior, and adapter_only=true
• pytest: OSEnvironmentBoundaryReceipt declares OS name, version, kernel version, and locale policy
• pytest: DependencyVersionEquivalenceReceipt declares all pinned dependency versions and their source hashes
• pytest: FloatingPointEnvironmentReceipt declares FP precision behavior, subnormal handling, and NaN policy
• pytest: EnvironmentDriftReceipt is produced on any deviation from declared environment
• pytest: CrossEnvironmentReplayReceipt verifies byte-identical output across declared equivalent environments
• pytest: CrossEnvironmentReplayProof aggregates all environment receipts into canonical proof hash
• pytest: no silent environment fallback
• pytest: no undeclared environment deviation
• pytest: no hardware authority claim
• pytest: IEEE 754 precision receipts from v178.5.x are required for any floating-point environment declaration
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: hardware environments are adapters, never authorities

Must Not Do

• no silent environment fallback
• no undeclared environment deviation
• no hardware authority claim
• no cross-environment claim without environment declaration
• no floating-point environment without IEEE 754 precision receipt
• no dependency version without pin manifest
• no OS environment without locale policy declaration
• no live hardware connections in tests

Dependency Boundaries

• depends on: v178.5.x IEEE 754 precision receipts (floating-point environment discipline)
• depends on: v170.x reproducible build receipts (build environment discipline)
• depends on: v177.x hardware abstraction receipts (hardware environment context)
• depends on: v188.x experiment scheduler receipts (scheduled cross-environment replay)
• feeds into: v191.x Reproducible Build / Hermetic Environment Receipts v2 (hermetic extension)
• feeds into: v192.x Global Proof Composition v2 (cross-environment replay binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v189.x implements the cross_environment_replay_receipt_hash entry in the terminal proof chain. It is identified as a Zenodo publication opportunity. It extends the cross-environment replay discipline of v148.6 and v151.8 into hardware and OS contexts, with explicit IEEE 754 floating-point environment receipts.

---

Phase: v190.x — Operator / IRC / TUI / CLI Federation Receipts v2

Status PLANNED

Source-Grounded Motivation v162.x established the IRC operator control surface. v175.x established the operator console unification. v190.x is the v2 federation extension that unifies IRC, TUI, and CLI operator surfaces under a single federated operator audit receipt.

Operator federation means: multiple operator surfaces may coexist, but every federated command must produce a receipt. Every federated session must declare: participating surfaces, command routing policy, conflict resolution policy, and audit trail policy. No hidden operator side-channels across federated surfaces. No autonomous execution authority.

The federated_operator_audit_receipt_hash in the terminal proof chain is anchored here.

Arc Reinterpretation

"Operator federation is a bounded multi-surface audit.

Every federated command is a receipt.
Every federated session is declared.
Conflict resolution is explicit.

No hidden cross-surface side-channels.
No autonomous execution authority.
The federation manifest is the contract.
The audit receipt is the proof."

Planned Releases

• v190.0 → FederatedOperatorManifest
• v190.1 → FederatedCommandReceipt
• v190.2 → SurfaceRegistrationReceipt
• v190.3 → CommandRoutingPolicyReceipt
• v190.4 → FederationConflictResolutionReceipt
• v190.5 → FederatedOperatorAuditReceipt
• v190.6 → FederationReplayEquivalenceReceipt

Expected Modules

• src/qec/analysis/federated_operator_manifest.py
• src/qec/analysis/federated_command_receipts.py
• src/qec/analysis/surface_registration_receipts.py
• src/qec/analysis/command_routing_policy_receipts.py
• src/qec/analysis/federation_conflict_resolution_receipts.py
• src/qec/analysis/federated_operator_audit_receipts.py
• src/qec/analysis/federation_replay_equivalence_receipts.py

Expected Artifacts

• FederatedOperatorManifest
• FederatedCommandReceipt
• SurfaceRegistrationReceipt
• CommandRoutingPolicyReceipt
• FederationConflictResolutionReceipt
• FederatedOperatorAuditReceipt
• FederationReplayEquivalenceReceipt

Expected Hashes

• federated_operator_manifest_hash (v190.0)
• federated_command_receipt_hash (v190.1)
• surface_registration_receipt_hash (v190.2)
• command_routing_policy_receipt_hash (v190.3)
• federation_conflict_resolution_receipt_hash (v190.4)
• federated_operator_audit_receipt_hash (v190.5)
• federation_replay_equivalence_receipt_hash (v190.6)

Core Rule

same federated operator manifest
+ same surface registration set
+ same command routing policy
+ same conflict resolution policy
→ same canonical federated operator state
→ same federated_operator_audit_receipt_hash

Acceptance Gates

• pytest: FederatedOperatorManifest declares all registered surfaces, routing policy, and conflict resolution policy
• pytest: FederatedCommandReceipt declares originating surface, command type, input hash, and output hash
• pytest: SurfaceRegistrationReceipt declares surface type (IRC/TUI/CLI), version, and local-only flag
• pytest: CommandRoutingPolicyReceipt declares routing rules, priority order, and fallback policy
• pytest: FederationConflictResolutionReceipt declares conflict type, resolution method, and winning surface
• pytest: FederatedOperatorAuditReceipt aggregates all federated command receipts into canonical audit hash
• pytest: FederationReplayEquivalenceReceipt verifies same command stream produces same federated state
• pytest: no hidden cross-surface side-channels
• pytest: no autonomous execution authority
• pytest: no live network calls in tests
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: operator surfaces are adapters, never authorities

Must Not Do

• no hidden cross-surface side-channels
• no autonomous execution authority
• no undeclared surface registration
• no command routing without policy declaration
• no conflict resolution without receipt
• no federated session without audit receipt
• no live network calls in tests

Dependency Boundaries

• depends on: v162.x IRC operator control surface
• depends on: v175.x operator console unification
• feeds into: v192.x Global Proof Composition v2 (federated operator binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v190.x implements the federated_operator_audit_receipt_hash entry in the terminal proof chain. It extends the operator surface discipline of v162.x and v175.x into the multi-surface federation domain, consistent with the Self-Hosting Reminder already present in the roadmap.

---

Phase: v191.x — Reproducible Build / Hermetic Environment Receipts v2

Status PLANNED

Source-Grounded Motivation v170.x established the reproducible build and supply-chain receipt discipline. v189.x established the cross-environment replay discipline. v191.x is the v2 hermetic extension that combines both into a full hermetic environment receipt.

A hermetic environment is a build and execution environment that is fully declared, fully pinned, and fully reproducible. Every hermetic environment must declare: build environment, execution environment, dependency pin manifest, network isolation policy, filesystem isolation policy, and replay equivalence contract. No undeclared external dependency. No silent network access. No filesystem side-effects outside declared scope.

The Debian Forky mandatory reproducibility milestone is cited as a research signal. v191.x implements the hermetic environment discipline that makes QEC builds and executions reproducible across environments.

The hermetic_environment_receipt_hash in the terminal proof chain is anchored here.

Arc Reinterpretation

"A hermetic environment is a declared contract.

The build environment is pinned.
The execution environment is pinned.
The network is isolated.
The filesystem is bounded.

No undeclared external dependency.
No silent network access.
No filesystem side-effects outside declared scope.

The hermetic receipt is the proof.
The environment is the adapter.
The receipt is the boundary."

Planned Releases

• v191.0 → HermeticEnvironmentManifest
• v191.1 → HermeticBuildReceipt
• v191.2 → HermeticExecutionReceipt
• v191.3 → NetworkIsolationPolicyReceipt
• v191.4 → FilesystemIsolationPolicyReceipt
• v191.5 → HermeticDependencyPinReceipt
• v191.6 → HermeticReplayEquivalenceReceipt
• v191.7 → HermeticEnvironmentReceipt

Expected Modules

• src/qec/analysis/hermetic_environment_manifest.py
• src/qec/analysis/hermetic_build_receipts.py
• src/qec/analysis/hermetic_execution_receipts.py
• src/qec/analysis/network_isolation_policy_receipts.py
• src/qec/analysis/filesystem_isolation_policy_receipts.py
• src/qec/analysis/hermetic_dependency_pin_receipts.py
• src/qec/analysis/hermetic_replay_equivalence_receipts.py
• src/qec/analysis/hermetic_environment_receipts.py

Expected Artifacts

• HermeticEnvironmentManifest
• HermeticBuildReceipt
• HermeticExecutionReceipt
• NetworkIsolationPolicyReceipt
• FilesystemIsolationPolicyReceipt
• HermeticDependencyPinReceipt
• HermeticReplayEquivalenceReceipt
• HermeticEnvironmentReceipt

Expected Hashes

• hermetic_environment_manifest_hash (v191.0)
• hermetic_build_receipt_hash (v191.1)
• hermetic_execution_receipt_hash (v191.2)
• network_isolation_policy_receipt_hash (v191.3)
• filesystem_isolation_policy_receipt_hash (v191.4)
• hermetic_dependency_pin_receipt_hash (v191.5)
• hermetic_replay_equivalence_receipt_hash (v191.6)
• hermetic_environment_receipt_hash (v191.7)

Core Rule

same hermetic environment manifest
+ same pinned build environment
+ same pinned execution environment
+ same network isolation policy
+ same filesystem isolation policy
+ same dependency pin manifest
→ same hermetic environment output
→ same hermetic_environment_receipt_hash

Acceptance Gates

• pytest: HermeticEnvironmentManifest declares all environment components and their receipt hashes
• pytest: HermeticBuildReceipt declares source hash, dependency pin manifest hash, and build environment hash
• pytest: HermeticExecutionReceipt declares execution environment hash, network isolation policy, and filesystem scope
• pytest: NetworkIsolationPolicyReceipt declares allowed network endpoints (empty list for fully hermetic)
• pytest: FilesystemIsolationPolicyReceipt declares allowed read/write paths and side-effect policy
• pytest: HermeticDependencyPinReceipt declares all dependencies with version, source hash, and checksum
• pytest: HermeticReplayEquivalenceReceipt verifies byte-identical output across hermetic replay runs
• pytest: HermeticEnvironmentReceipt aggregates all hermetic receipts into canonical environment hash
• pytest: no undeclared external dependency
• pytest: no silent network access
• pytest: no filesystem side-effects outside declared scope
• pytest: reproducible build receipts from v170.x are required inputs
• pytest: cross-environment replay receipts from v189.x are required inputs
• dependency boundary: no import from src/qec/decoder/
• dependency boundary: build systems and execution environments are adapters, never authorities

Must Not Do

• no undeclared external dependency
• no silent network access
• no filesystem side-effects outside declared scope
• no hermetic claim without hermetic receipt
• no dependency without pin receipt
• no build without hermetic build receipt
• no execution without hermetic execution receipt
• no installer without checksum verification

Dependency Boundaries

• depends on: v170.x reproducible build receipts (build discipline)
• depends on: v189.x cross-environment replay receipts (environment equivalence discipline)
• feeds into: v192.x Global Proof Composition v2 (hermetic environment binding)
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v191.x implements the hermetic_environment_receipt_hash entry in the terminal proof chain. It is identified as a Zenodo publication opportunity. It combines the reproducible build discipline of v170.x with the cross-environment replay discipline of v189.x into a full hermetic environment receipt, consistent with the Reproducible Build Reminder already present in the roadmap.

---

Phase: v192.x — Global Proof Composition v2

Status PLANNED

Source-Grounded Motivation v161.x established the global validation index, global threshold contract, global truth receipt, and global replay proof. These form the v1 global proof composition. v192.x is the v2 global proof composition that registers all post-v173 receipts and produces a canonical global_proof_composition_v2_hash.

v192.x is the terminal arc of the roadmap. It does not introduce new proof concepts. It registers, binds, and canonicalizes all receipts produced by v167.x through v191.x into a single, replay-safe, hash-bound global proof composition artifact.

The global proof composition v2 is not a truth claim. It is a bounded registry of declared, receipt-bound artifacts. Every registered artifact has a declared source, a declared claim scope, and a declared review status. No artifact may expand its claim scope at composition time. No artifact may claim truth beyond its declared evidence scope.

v192.x is a major structural arc. It is identified as a Zenodo publication opportunity.

Arc Reinterpretation

"Global proof composition v2 is a bounded registry.

It registers receipts.
It does not generate truth.
It does not expand claim scopes.
It does not grant execution authority.

Every registered artifact has:
  - declared source
  - declared claim scope
  - declared review status
  - stable hash

The composition is the index.
The receipt is the proof.
The hash is the anchor."

Planned Releases

• v192.0 → PostV173ArtifactRegistry
• v192.1 → SourceClaimCompositionReceipt
• v192.2 → BenchmarkAndHardwareBoundaryCompositionReceipt
• v192.3 → OperatorAndReproducibilityCompositionReceipt
• v192.4 → SymbolicAndInterpretabilityCompositionReceipt
• v192.5 → GlobalProofCompositionV2

Expected Modules

• src/qec/analysis/post_v173_artifact_registry.py
• src/qec/analysis/source_claim_composition_receipts.py
• src/qec/analysis/benchmark_hardware_boundary_composition_receipts.py
• src/qec/analysis/operator_reproducibility_composition_receipts.py
• src/qec/analysis/symbolic_interpretability_composition_receipts.py
• src/qec/analysis/global_proof_composition_v2.py

Expected Artifacts

• PostV173ArtifactRegistry
• SourceClaimCompositionReceipt
• BenchmarkAndHardwareBoundaryCompositionReceipt
• OperatorAndReproducibilityCompositionReceipt
• SymbolicAndInterpretabilityCompositionReceipt
• GlobalProofCompositionV2

Expected Hashes

• post_v173_artifact_registry_hash (v192.0)
• source_claim_composition_receipt_hash (v192.1)
• benchmark_hardware_boundary_composition_receipt_hash (v192.2)
• operator_reproducibility_composition_receipt_hash (v192.3)
• symbolic_interpretability_composition_receipt_hash (v192.4)
• global_proof_composition_v2_hash (v192.5)

Core Rule

same post-v173 artifact registry
+ same source claim receipts
+ same benchmark boundary receipts
+ same operator/reproducibility receipts
+ same symbolic/interpretability receipts
+ same declared composition policy
→ same global_proof_composition_v2_hash

Composition Registry

The PostV173ArtifactRegistry must include, at minimum, the following receipt families:

qudit_replay_equivalence_receipt_hash           (v167.x)
telemetry_replay_equivalence_receipt_hash       (v168.x)
symbolic_sandbox_replay_receipt_hash            (v169.x)
build_replay_equivalence_receipt_hash           (v170.x)
knowledge_base_replay_equivalence_receipt_hash  (v171.x)
materials_replay_boundary_receipt_hash          (v172.x)
game_world_replay_equivalence_receipt_hash      (v173.x)
bp_replay_equivalence_receipt_hash              (v174.x)
console_replay_equivalence_receipt_hash         (v175.x)
syndrome_stream_replay_equivalence_receipt_hash (v176.x)
control_plane_replay_equivalence_receipt_hash   (v177.x)
resource_replay_equivalence_receipt_hash        (v178.x)
contextuality_replay_equivalence_receipt_hash   (v179.x)
qml_replay_equivalence_receipt_hash             (v180.x)
tool_dispatch_replay_proof_hash                 (v181.x)
interpretability_replay_equivalence_receipt_hash (v182.x)
publication_replay_equivalence_receipt_hash     (v183.x)
benchmark_replay_equivalence_receipt_hash       (v184.x)
materials_signal_boundary_receipt_v2_hash       (v185.x)
symbolic_compiler_manifest_hash                 (v186.x)
audit_trail_receipt_hash                        (v187.x)
experiment_scheduler_replay_proof_hash          (v188.x)
cross_environment_replay_receipt_hash           (v189.x)
federated_operator_audit_receipt_hash           (v190.x)
hermetic_environment_receipt_hash               (v191.x)

Acceptance Gates

• pytest: PostV173ArtifactRegistry declares all registered receipt families and their hash chain
• pytest: SourceClaimCompositionReceipt verifies all source claims are within declared evidence scope
• pytest: BenchmarkAndHardwareBoundaryCompositionReceipt verifies all benchmark claims have comparator receipts
• pytest: OperatorAndReproducibilityCompositionReceipt verifies all operator receipts and hermetic receipts
• pytest: SymbolicAndInterpretabilityCompositionReceipt verifies all symbolic claims have claim_mode = SYMBOLIC_ONLY
• pytest: GlobalProofCompositionV2 is hash-stable across PYTHONHASHSEED values
• pytest: GlobalProofCompositionV2 rejects any artifact without declared claim scope
• pytest: GlobalProofCompositionV2 rejects any artifact without declared review status
• pytest: GlobalProofCompositionV2 rejects any artifact with claim scope expansion at composition time
• pytest: no global truth expansion beyond registered receipts
• pytest: no semantic truth claims
• pytest: no hardware authority claims
• pytest: no symbolic claim drift
• pytest: no benchmark marketing claims
• dependency boundary: no import from src/qec/decoder/

Must Not Do

• no global truth expansion beyond registered receipts
• no semantic truth claims
• no hardware authority claims
• no symbolic claim drift
• no benchmark marketing claims
• no claim scope expansion at composition time
• no artifact registration without declared source
• no artifact registration without declared review status
• no composition without replay equivalence proof

Dependency Boundaries

• depends on: all v167.x through v191.x receipt families (complete post-v173 arc)
• depends on: v161.x global validation index (v1 composition baseline)
• terminal arc: no further phases depend on v192.x
• does not modify: src/qec/decoder/

Relationship to Existing Roadmap v192.x implements the global_proof_composition_v2_hash terminal entry in the proof chain. It is identified as a Zenodo publication opportunity. It extends the global proof composition discipline of v161.x to cover all post-v173 receipt families. The existing placeholder already declares the core rule and must-not-do constraints; this section completes it with full module, artifact, hash, acceptance gate, and composition registry structure.

---

Updated Terminal Proof Chain (v167 → v192 Extension)

decoder_promotion_receipt_hash                  (v166.8)
→ qldpc_construction_receipt_hash               (v166.x)
→ qudit_dimension_policy_manifest_hash          (v167.0)
→ qudit_stabilizer_generator_receipt_hash       (v167.1)
→ qudit_syndrome_extraction_receipt_hash        (v167.2)
→ qudit_equivalence_class_receipt_hash          (v167.3)
→ qudit_code_parameter_receipt_hash             (v167.4)
→ qudit_replay_equivalence_receipt_hash         (v167.5)
→ qudit_benchmark_boundary_receipt_hash         (v167.6)
→ proof_telemetry_manifest_hash                 (v168.0)
→ midi_event_stream_receipt_hash                (v168.1)
→ pcm_render_spec_receipt_hash                  (v168.2)
→ sonification_projection_receipt_hash          (v168.3)
→ telemetry_replay_equivalence_receipt_hash     (v168.4)
→ symbolic_grammar_manifest_hash                (v169.0)
→ symbolic_hypothesis_receipt_hash              (v169.1)
→ mathematical_identity_candidate_receipt_hash  (v169.2)
→ cosmological_claim_boundary_receipt_hash      (v169.3)
→ symbolic_sandbox_replay_receipt_hash          (v169.4)
→ reproducible_build_manifest_hash              (v170.0)
→ dependency_pin_manifest_hash                  (v170.1)
→ sbom_receipt_hash                             (v170.2)
→ installer_provenance_receipt_hash             (v170.3)
→ build_environment_declaration_receipt_hash    (v170.4)
→ build_replay_equivalence_receipt_hash         (v170.5)
→ build_environment_drift_receipt_hash          (v170.6)
→ knowledge_base_manifest_hash                  (v171.0)
→ agent_memory_write_receipt_hash               (v171.1)
→ agent_memory_read_receipt_hash                (v171.2)
→ memory_promotion_receipt_hash                 (v171.3)
→ knowledge_base_replay_equivalence_receipt_hash (v171.4)
→ memory_eviction_policy_receipt_hash           (v171.5)
→ materials_signal_manifest_hash                (v172.0)
→ graphene_signal_receipt_hash                  (v172.1)
→ photonic_signal_receipt_hash                  (v172.2)
→ diamond_siv_signal_receipt_hash               (v172.3)
→ materials_memory_claim_boundary_receipt_hash  (v172.4)
→ fiber_optic_signal_receipt_hash               (v172.5)
→ hardware_library_provenance_receipt_hash      (v172.6)
→ materials_replay_boundary_receipt_hash        (v172.7)
→ interactive_proof_world_manifest_hash         (v173.0)
→ citizen_science_observation_receipt_hash      (v173.1)
→ participant_contribution_boundary_receipt_hash (v173.2)
→ contribution_aggregation_receipt_hash         (v173.3)
→ game_world_replay_equivalence_receipt_hash    (v173.4)
→ citizen_science_proof_receipt_hash            (v173.5)
→ bp_dynamics_manifest_hash                     (v174.0)
→ bp_iteration_policy_receipt_hash              (v174.1)
→ fixed_point_trap_receipt_hash                 (v174.2)
→ attractor_basin_boundary_receipt_hash         (v174.3)
→ trapping_set_characterization_receipt_hash    (v174.4)
→ bp_phase_space_receipt_hash                   (v174.5)
→ bp_replay_equivalence_receipt_hash            (v174.6)
→ operator_console_manifest_hash                (v175.0)
→ console_command_receipt_hash                  (v175.1)
→ console_display_state_receipt_hash            (v175.2)
→ operator_session_receipt_hash                 (v175.3)
→ console_replay_equivalence_receipt_hash       (v175.4)
→ syndrome_stream_manifest_hash                 (v176.0)
→ streaming_window_receipt_hash                 (v176.1)
→ syndrome_replay_receipt_hash                  (v176.2)
→ stream_buffer_policy_receipt_hash             (v176.3)
→ syndrome_ordering_receipt_hash                (v176.4)
→ stream_latency_boundary_receipt_hash          (v176.5)
→ syndrome_stream_replay_equivalence_receipt_hash (v176.6)
→ hardware_abstraction_manifest_hash            (v177.0)
→ control_plane_protocol_receipt_hash           (v177.1)
→ hardware_adapter_boundary_receipt_hash        (v177.2)
→ latency_envelope_receipt_hash                 (v177.3)
→ fidelity_model_boundary_receipt_hash          (v177.4)
→ noise_model_source_receipt_hash               (v177.5)
→ control_plane_replay_equivalence_receipt_hash (v177.6)
→ distillation_overhead_receipt_hash            (v178.0)
→ logical_cycle_overhead_receipt_hash           (v178.1)
→ resource_budget_replay_receipt_hash           (v178.2)
→ physical_qubit_count_receipt_hash             (v178.3)
→ gate_synthesis_overhead_receipt_hash          (v178.4)
→ error_budget_allocation_receipt_hash          (v178.5)
→ resource_accounting_manifest_hash             (v178.6)
→ resource_replay_equivalence_receipt_hash      (v178.7)
→ ieee754_precision_format_manifest_hash        (v178.5.0)
→ safe_bit_reinterpretation_policy_hash         (v178.5.1)
→ fast_approximation_receipts_hash              (v178.5.2)
→ fast_inv_sqrt_receipt_hash                    (v178.5.2)
→ fast_exp_receipt_hash                         (v178.5.2)
→ sign_bit_operation_receipts_hash              (v178.5.3)
→ ulp_epsilon_receipt_hash                      (v178.5.4)
→ float_ordering_receipt_hash                   (v178.5.5)
→ reduced_precision_adapter_receipt_hash        (v178.5.6)
→ float_integer_test_receipt_hash               (v178.5.7)
→ hardware_float_adapter_boundary_hash          (v178.5.8)
→ contextuality_threshold_receipt_hash          (v179.0)
→ topological_boundary_receipt_hash             (v179.1)
→ geometry_replay_validation_receipt_hash       (v179.2)
→ code_switching_boundary_receipt_hash          (v179.3)
→ topological_code_manifest_hash                (v179.4)
→ contextuality_replay_equivalence_receipt_hash (v179.5)
→ qml_boundary_manifest_hash                    (v180.0)
→ circuit_sampling_policy_receipt_hash          (v180.1)
→ qml_precision_boundary_receipt_hash           (v180.2)
→ qml_benchmark_boundary_receipt_hash           (v180.3)
→ qml_replay_equivalence_receipt_hash           (v180.4)
→ quantum_advantage_claim_boundary_receipt_hash (v180.5)
→ tool_dispatch_manifest_hash                   (v181.0)
→ tool_call_receipt_hash                        (v181.1)
→ dispatch_decision_receipt_hash                (v181.2)
→ recursive_dispatch_boundary_receipt_hash      (v181.3)
→ tool_output_boundary_receipt_hash             (v181.4)
→ agent_execution_context_receipt_hash          (v181.5)
→ tool_dispatch_replay_proof_hash               (v181.6)
→ local_only_execution_boundary_receipt_hash    (v181.7)
→ interpretability_boundary_manifest_hash       (v182.0)
→ sparse_feature_extraction_receipt_hash        (v182.1)
→ feature_activation_boundary_receipt_hash      (v182.2)
→ interpretability_claim_scope_receipt_hash     (v182.3)
→ interpretability_replay_equivalence_receipt_hash (v182.4)
→ reproducible_publication_manifest_hash        (v183.0)
→ figure_data_source_receipt_hash               (v183.1)
→ table_computation_receipt_hash                (v183.2)
→ claim_evidence_scope_receipt_hash             (v183.3)
→ publication_human_review_receipt_hash         (v183.4)
→ citation_chain_integrity_receipt_hash         (v183.5)
→ publication_replay_equivalence_receipt_hash   (v183.6)
→ zenodo_provenance_receipt_hash                (v183.7)
→ benchmark_ladder_manifest_hash                (v184.0)
→ comparator_identity_receipt_hash              (v184.1)
→ benchmark_corpus_receipt_hash                 (v184.2)
→ hardware_benchmark_declaration_receipt_hash   (v184.3)
→ benchmark_measurement_receipt_hash            (v184.4)
→ external_comparator_boundary_receipt_hash     (v184.5)
→ benchmark_claim_scope_receipt_hash            (v184.6)
→ benchmark_replay_equivalence_receipt_hash     (v184.7)
→ benchmark_ladder_receipt_hash                 (v184.8)
→ materials_signal_manifest_v2_hash             (v185.0)
→ integrated_photonic_circuit_receipt_hash      (v185.1)
→ cv_optics_signal_receipt_hash                 (v185.2)
→ qkd_signal_receipt_hash                       (v185.3)
→ device_stack_provenance_receipt_hash          (v185.4)
→ materials_memory_claim_boundary_receipt_v2_hash (v185.5)
→ materials_signal_boundary_receipt_v2_hash     (v185.6)
→ symbolic_tokenizer_receipt_v2_hash            (v186.0)
→ diagram_compiler_manifest_v2_hash             (v186.1)
→ symbolic_boundary_replay_receipt_v2_hash      (v186.2)
→ topological_diagram_receipt_hash              (v186.3)
→ grammar_extension_receipt_hash                (v186.4)
→ reproducible_figure_receipt_hash              (v186.5)
→ symbolic_compiler_manifest_hash               (v186.6)
→ human_audit_manifest_hash                     (v187.0)
→ audit_finding_receipt_hash                    (v187.1)
→ red_team_scope_receipt_hash                   (v187.2)
→ red_team_finding_receipt_hash                 (v187.3)
→ remediation_contract_receipt_hash             (v187.4)
→ audit_trail_receipt_hash                      (v187.5)
→ audit_replay_equivalence_receipt_hash         (v187.6)
→ experiment_scheduler_manifest_hash            (v188.0)
→ scheduled_experiment_receipt_hash             (v188.1)
→ feedback_loop_stability_receipt_hash          (v188.2)
→ real_time_boundary_receipt_hash               (v188.3)
→ transport_backpressure_receipt_hash           (v188.4)
→ collapse_prevention_signal_receipt_hash       (v188.5)
→ experiment_dependency_receipt_hash            (v188.6)
→ experiment_scheduler_replay_proof_hash        (v188.7)
→ cross_environment_replay_manifest_hash        (v189.0)
→ environment_declaration_receipt_hash          (v189.1)
→ hardware_environment_boundary_receipt_hash    (v189.2)
→ os_environment_boundary_receipt_hash          (v189.3)
→ dependency_version_equivalence_receipt_hash   (v189.4)
→ floating_point_environment_receipt_hash       (v189.5)
→ environment_drift_receipt_hash                (v189.6)
→ cross_environment_replay_receipt_hash         (v189.7)
→ cross_environment_replay_proof_hash           (v189.8)
→ federated_operator_manifest_hash              (v190.0)
→ federated_command_receipt_hash                (v190.1)
→ surface_registration_receipt_hash             (v190.2)
→ command_routing_policy_receipt_hash           (v190.3)
→ federation_conflict_resolution_receipt_hash   (v190.4)
→ federated_operator_audit_receipt_hash         (v190.5)
→ federation_replay_equivalence_receipt_hash    (v190.6)
→ hermetic_environment_manifest_hash            (v191.0)
→ hermetic_build_receipt_hash                   (v191.1)
→ hermetic_execution_receipt_hash               (v191.2)
→ network_isolation_policy_receipt_hash         (v191.3)
→ filesystem_isolation_policy_receipt_hash      (v191.4)
→ hermetic_dependency_pin_receipt_hash          (v191.5)
→ hermetic_replay_equivalence_receipt_hash      (v191.6)
→ hermetic_environment_receipt_hash             (v191.7)
→ post_v173_artifact_registry_hash              (v192.0)
→ source_claim_composition_receipt_hash         (v192.1)
→ benchmark_hardware_boundary_composition_receipt_hash (v192.2)
→ operator_reproducibility_composition_receipt_hash (v192.3)
→ symbolic_interpretability_composition_receipt_hash (v192.4)
→ global_proof_composition_v2_hash              (v192.5)