@journeykit/canon 0.12.0 → 0.13.1

package/dist/skills/SKILL.src.md

@@ -11,17 +11,19 @@ Use this single skill when working with Canon in a consumer repository. Canon's
 
 ```bash
 canon init
-canon skills install                         # detects .claude/.codex/.omp/.pi; if none exists, installs Claude+Codex
+canon skills install                         # detects .claude/.agents|.codex/.omp/.pi; if none exists, installs Claude+Codex
 canon skills install --providers=claude,codex,omp,pi
 canon skills check                           # read-only drift check
 canon skills doctor                          # diagnostics; never deletes user files
 ```
 
 The canonical bundle is read-only. Claude receives
-`.claude/skills/canon/SKILL.md`; Codex receives the flattened
-`.codex/skills/canon.md`; OMP receives `.omp/skills/canon/SKILL.md`; and Pi
-receives `.pi/skills/canon/SKILL.md`. Every selected provider receives matching
-`reference/**` and `scripts/**` sidecars under its `canon` bundle. OMP and Pi
+`.claude/skills/canon/SKILL.md`; Codex receives
+`.agents/skills/canon/SKILL.md`; OMP receives `.omp/skills/canon/SKILL.md`; and
+Pi receives `.pi/skills/canon/SKILL.md`. Every selected provider receives
+matching `reference/**` and `scripts/**` sidecars under its `canon` bundle. A
+legacy `.codex/skills/canon*` projection from canon 0.13.0 or earlier is
+migrated by `canon skills install` and reported by `check`/`doctor`. OMP and Pi
 projections are project-local passive skill bundles: the
 `scripts/canon-retrieve-pre-dispatch.sh` file is available as a sidecar, not
 installed as a native hook. Load only the reference needed for the current

package/dist/skills/reference/canon-context.md

@@ -100,7 +100,27 @@ output form includes captured prompt bytes.
 - **`joinKeys`** — the join-spine grammar string per key (`regime_key`,
   `session_id`, `task_id`, …).
 - **`policy`** — the resolved `policy.yaml`-derived evidence/trust
-  requirements per kind.
+  requirements per kind, plus `clean` and the policy `diagnostics`
+  (an unknown top-level key is one). In `--json`, `spec_coverage` and
+  `evidence_binding` are objects keyed like `policy.yaml`, with booleans
+  as booleans and lists as arrays (an empty `scope`/`case`/`lane` means
+  unfiltered):
+
+  ```json
+  "spec_coverage": {"require_evidence": true, "scope": ["building", "verifying"],
+    "exclude_lanes": ["process"], "require_cases": ["failure"],
+    "require_review": {"scope": ["verifying", "shipped"], "distinct_actor": true, "block_on_findings": true}},
+  "evidence_binding": {"mode": "warn", "strength": "report", "case": ["failure"], "lane": [], "scope": []}
+  ```
+
+  `spec_coverage` is `null` when absent, `require_review` is `null` when
+  its sub-section is absent, and `evidence_binding` is omitted when absent.
+  A present but unusable section is `{"invalid": "<detail>"}`. The text
+  outline keeps its one-line summaries (`spec_coverage: INVALID — …`).
+- **`review`** — finding severities and dispositions, the Review fields,
+  and `requireReview`: `{"scope": [...], "distinctActor": true,
+  "blockOnFindings": true}`, or `null` when review is not required
+  (absent, or `spec_coverage` is absent or invalid).
 - **`vocab`** — the typed authoring vocabulary (directive/enum/evidence-
   kind index) + its content-hash version.
 - **`cel`** — the per-kind CEL binding surface: the bindable

package/dist/skills/reference/canon-policy.md

@@ -6,6 +6,23 @@ mapping guards, retrieval filters, and a template's `applies_when:`
 section. Wherever a `policy.yaml` predicate or an `applies_when:` string
 appears, it is a CEL expression evaluated against the target record.
 
+## Top-level keys
+
+`.canon/policy.yaml` accepts exactly these top-level keys:
+`adapter_capabilities`, `allowed_signers`, `approval`, `experimental`,
+`query`, `risk_approvals`, `risk_routing`, `risk_tiers`, `schema`,
+`spec_coverage`, `staleness`, `trust_required`, `trust_sample`
+(`risk_approvals` and a top-level `allowed_signers` are compatibility
+aliases for `approval.allowed_signers`). Any other key, usually a typo,
+is a policy diagnostic, and `canon gate check` fails on it:
+
+```
+uncovered-cell spec_coverag — `spec_coverag` is not a known top-level policy.yaml key (known keys: adapter_capabilities, allowed_signers, approval, experimental, query, risk_approvals, risk_routing, risk_tiers, schema, spec_coverage, staleness, trust_required, trust_sample); fix it or remove it — an unknown key is never silently ignored
+```
+
+The known keys still resolve normally, and `canon context` lists the
+diagnostic and reports `clean: false`.
+
 ## The closed CEL profile
 
 Every expression is evaluated against exactly one variable, `record` (the

package/package.json

@@ -1,6 +1,6 @@
 {
   "name": "@journeykit/canon",
-  "version": "0.12.0",
+  "version": "0.13.1",
   "description": "canon — harness knowledge substrate. JS launcher: platform detection + native binary resolution.",
   "license": "MIT",
   "type": "module",
@@ -30,9 +30,9 @@
   },
   "dependencies": {},
   "optionalDependencies": {
-    "@journeykit/canon-core-darwin-arm64": "0.12.0",
-    "@journeykit/canon-core-linux-x64": "0.12.0",
-    "@journeykit/canon-core-win32-x64": "0.12.0"
+    "@journeykit/canon-core-darwin-arm64": "0.13.1",
+    "@journeykit/canon-core-linux-x64": "0.13.1",
+    "@journeykit/canon-core-win32-x64": "0.13.1"
   },
   "devDependencies": {
     "@types/node": "^20.0.0",