chore(ech05): regression fixture → JCS, stage 6 first target, #431 row, owner norms [stack 3/3]

.claude/CLAUDE.md

@@ -152,3 +152,32 @@ package definitions (not metadata files) and must NOT be deleted.
 - SHA-pinned dependencies
 - SPDX license headers on all files
 
+## Owner norms (2026-10-14)
+
+These come from the owner. The format rules point at their normative
+source in `hyperpolymath/standards`; read that, not a paraphrase.
+
+### Asking the owner
+
+- Ask decisions with the selector tool (structured options, the
+  recommended one first, with what each option implies). Use prose only
+  for reports and explanations.
+- Work lands as **stacked PRs**: small, each independently reviewable.
+- Fix a finding at its source before reaching for a baseline or ledger.
+  The governance exemption ratchet rejects ledger growth unless a commit
+  or the PR body carries a `Ratchet-exception:` line naming that ledger.
+
+### Formats (standard, estate-wide)
+
+| Kind | Standard | Source |
+|------|----------|--------|
+| YAML | KYAML, formatted with standards `scripts/kyaml-format.sh` (mikefarah yq v4 with the KYAML encoder). Estate-owned config uses `.kyaml` unless a tool fixes the name. No Python YAML parser. | `3-practice/YAML-POLICY.adoc` §3.4 (D280, D307, D293) |
+| Workflows | KYAML too, but only through the step 6 gate in `standards#1025` (Dependabot / `gh actions-lock` drift accepted in writing, lock coverage, `actionlint`, a run proof per workflow). Lockfiles such as `actions.lock` are tool-owned: never hand-edit or reformat them. | `YAML-POLICY.adoc` §5–§6 |
+| JSON | I-JSON (RFC 7493) serialised as JCS (RFC 8785), plus exactly one final LF. Extensions do not change. JSONL is one JCS object per line. JSONC files (`devcontainer.json`, `.vscode/`) are exempt. | `3-practice/JSON-POLICY.adoc` (D306) |
+| UUID | Current policy: UUIDv7 for records and events, and UUIDv8 over JCS content bytes for content IDs. Keep this split until the corresponding `ids.rs` migration lands. Apply the Profile T/C rule and prohibit new v4/v7 minting after that migration. echidna mints only through `src/rust/ids.rs`. | `docs/decisions/ADR-008-uuid-v8-estate-standard.adoc` (D305, D305b, D320) |
+
+Keep another format only where a consumer the owner does not control
+would break (for example a tool that rewrites the file). Record why
+beside the file. Tools the owner controls are changed to handle the
+standard instead of being exempted.
+

docs/ROADMAP.adoc

@@ -212,9 +212,10 @@ kernel‑checked translation. `+ResultArbiter+` grades each verdict
 (ECH‑05), so a synonym collision cannot reach `+sound+`; dispatch attaches
 no evidence yet, so every dispatched verdict is `+belief+`
 |*A `+KernelTranslation+` certificate is producible for a named backend
-pair*, and a verdict on that pair reaches `+sound+`. First candidate:
-HOL4 ↔ HOL Light through OpenTheory (6a); then Lean 4 ↔ Rocq through
-Dedukti (6b). Pair table: `+docs/design/ADVANCED_TYPING_GAP_ANALYSIS.adoc+`
+pair*, and a verdict on that pair reaches `+sound+`. First target (owner
+decision, 2026-10-14): HOL4 ↔ HOL Light through OpenTheory (6a); next
+candidate Lean 4 ↔ Rocq through Dedukti (6b). Pair table:
+`+docs/design/ADVANCED_TYPING_GAP_ANALYSIS.adoc+`
 § "Backend pairs that can share a term"
 |===
 

docs/design/ADVANCED_TYPING_GAP_ANALYSIS.adoc

@@ -244,7 +244,7 @@ certificate for one pair is the ROADMAP stage 6 acceptance.
 |Backend pair |`ExchangeFormat` |External route |In ECHIDNA
 |HOL4 ↔ HOL Light |`open_theory` |OpenTheory articles, both directions;
 checked by the OpenTheory article reader |DESIGNED (`opentheory.rs`
-exports and imports proof states, not checked articles); first candidate
+exports and imports proof states, not checked articles); first target (2026-10-14)
 |Lean 4 ↔ Rocq |`dedukti` / `lambdapi` |Lean and Coq encodings in λΠ
 modulo (Lean4Less/lean4dk, CoqInE); checked by Dedukti or Lambdapi
 |DESIGNED (`dedukti.rs`, `lambdapi.rs` translate syntax only)

issues/README.md

@@ -20,7 +20,7 @@ awk 'f;/^# /{f=1}' issues/<file> | sed '1{/^$/d}' > /tmp/body.md
 | ECH-02 | hyperpolymath/echidna        | #421         | `echidna-02-cmaes-julia-gnn.md`                       |
 | ECH-03 | hyperpolymath/echidna        | #422         | `echidna-03-chapel-island-migration.md`               |
 | ECH-04 | hyperpolymath/echidna        | #424         | `echidna-04-ffi-abi-bridge.md`                        |
-| ECH-05 | hyperpolymath/echidna        | not filed    | `echidna-05-fidelity-evidence.md`                     |
+| ECH-05 | hyperpolymath/echidna        | #431         | `echidna-05-fidelity-evidence.md`                     |
 | PB-01  | hyperpolymath/proof-burrower | not filed    | `proof-burrower-01-combinatorial-tactic-synthesis.md` |
 | PB-02  | hyperpolymath/proof-burrower | not filed    | `proof-burrower-02-ledger-fitness-formulation.md`     |
 | EB-01  | hyperpolymath/echidnabot     | not filed    | `echidnabot-01-adaptive-ci-portfolio-scheduler.md`    |
@@ -30,7 +30,7 @@ awk 'f;/^# /{f=1}' issues/<file> | sed '1{/^$/d}' > /tmp/body.md
 ## Status
 
 - **ECH-01..04** are filed on hyperpolymath/echidna and carry all their labels. On 2026-10-09 their bodies were replaced by hand with a different text that is not these files; the owner decides which is canonical (see `docs/handover/PARADISEO-ROADMAP-PROMPT.adoc`, WP0). `scripts/issues/sync-roadmap-issues.sh` syncs the bodies from these files (dry run by default; refuses to replace hand-edited bodies without `--overwrite`).
-- **ECH-05** is a draft for hyperpolymath/echidna, not filed. Its implementation (fidelity evidence on `ProverAttempt`, `level` on `ArbitratedVerdict`, `EchidnaABI.Arbitration`, the fidelity tests) lands with the PR that adds the file.
+- **ECH-05** is filed as #431 and closed as completed by #430 (fidelity evidence on `ProverAttempt`, `level` on `ArbitratedVerdict`, `EchidnaABI.Arbitration`, the fidelity tests). It is the arbiter-typing slice of #429, which stays open for one-kernel (Agda) ingestion of fibre coverage, round-trip and cross-prover agreement.
 - **PB-01, PB-02, EB-01** are drafts for copying to their own repositories. They are not filed.
 
 ## Provenance of the PB and EB drafts
@@ -49,6 +49,6 @@ Labels named in the drafts that did not exist when they were written (checked 20
 |------------------------------|---------------------------------------------------------------------|
 | hyperpolymath/proof-burrower | `area:swarm`, `rust`, `search-tactics`, `area:ledger`, `heuristics` |
 | hyperpolymath/echidnabot     | `area:dispatcher`, `ci-optimization` (`rust` exists)                |
-| hyperpolymath/echidna        | `area:arbitration`, `area:exchange`, `epistemic-types`, `residual-evidence-types` (ECH-05; checked 2026-10-14) |
+| hyperpolymath/echidna        | none on the repo since #431 created `area:arbitration`, `area:exchange`, `epistemic-types`, `residual-evidence-types` (default colour, no description); they are not yet in the estate label source (`.github/labels.json` is generated from hyperpolymath/.git-private-farm) |
 
-All labels used by ECH-01..04 exist on hyperpolymath/echidna; ECH-05's `enhancement` and `idris2` do.
+All labels used by ECH-01..05 exist on hyperpolymath/echidna.

tests/arbitration_regression_test.rs

@@ -16,6 +16,10 @@
 //! the ECH-05 additions, and every replayed verdict must sit at level
 //! `belief` — the attempts carry no fidelity evidence.
 //!
+//! The fixture is JCS (RFC 8785) I-JSON, the estate format standard;
+//! `fixture_is_jcs_canonical` keeps it that way. It was re-serialised to
+//! JCS after generation without changing any value.
+//!
 //! Regenerate only at a commit whose numbers are the intended reference:
 //!
 //! ```sh
@@ -296,7 +300,21 @@ fn regenerate_arbitration_fixture() {
         "tolerance": TOLERANCE,
         "cases": replay(),
     });
-    let mut out = serde_json::to_string_pretty(&doc).unwrap();
+    // Estate format standard: JCS (RFC 8785) I-JSON, plus the final LF
+    // that .editorconfig requires.
+    let mut out = serde_json_canonicalizer::to_string(&doc).unwrap();
     out.push('\n');
     std::fs::write(FIXTURE, out).expect("write fixture");
 }
+
+#[test]
+fn fixture_is_jcs_canonical() {
+    let raw = std::fs::read_to_string(FIXTURE).expect("fixture present");
+    let body = raw.strip_suffix('\n').expect("fixture ends with one LF");
+    let value: Value = serde_json::from_str(body).expect("fixture is JSON");
+    assert_eq!(
+        body,
+        serde_json_canonicalizer::to_string(&value).unwrap(),
+        "fixture must be JCS (RFC 8785) canonical"
+    );
+}