fix(ci): green the trunk — dangerous-pattern excludes, markdownlint, skill-flow danglers, derived-artifact regen, stale bats (ag-ws4cl)

Trunk on main has been RED since 2026-06-08 (last green c9c3e270). Restores
every blocking gate to green on a branch. Five fix categories:

1. security / dangerous-pattern: add install-claude.sh, install-agy.sh, and
   install.sh to the curl|bash --exclude list in validate.yml. install.sh
   gained an AGY install-instructions echo (curl|bash) on 2026-06-09; the two
   new install scripts mirror the already-excluded install-codex/opencode set.

2. lint / markdownlint: fix 3 pre-existing failures without deleting content —
   MD037 (backtick the glob paths so * isn't read as emphasis) in
   dirty-main-attribution; two MD004 (rewrap so a continuation '+' isn't read
   as a plus-style bullet) in ag-s43tg-resume + skill-prune-phase2.

3. skill-flow danglers: the Phase-2 skill prune (40ee1c34f) deleted
   brainstorm/design/complexity/ratchet/audit-report/learning/
   operating-loop-skill/cass-memory but left consumes/metadata.dependencies/
   context_rel edges pointing at them. Removed the dangling edges from
   discovery, refactor, eval-outcomes, rpi, operationalize, agy-native, perf,
   test, acfs. converge (thin memo) now consumes the whitelisted command-help
   artifact; converge + beads-bv (orphaned when operating-loop-skill was pruned)
   added to skill-flow-standalone.txt with rationale.

4. derived-artifact drift: regenerated catalog.json, registry.json,
   context-map.md, and the 10 skills-codex .agentops-generated.json twins +
   manifest via scripts/regen-all.sh after the frontmatter edits.

5. stale bats: brainstorm-discovery-ideation.bats followed the deliberate
   bd->br tracker migration (f650d41bb) — assert br / ban bd — and the codex
   anti-leak guard now targets the real codex twin (skills-codex/discovery)
   instead of the shared Claude reference (which legitimately uses
   AskUserQuestion). install-sh-runtime-detection.bats updated to the new
   multi-vendor install-suggestion output format.

Did NOT touch tests/scripts/validate-skill-disposition-schema.bats: its
"bad revision origin/main" failure is a CI-env artifact (checkout lacks the
origin/main named ref); all 24 cases pass locally with the ref present and the
4 guarded files unchanged.

Co-Authored-By: Claude Opus 4.8 <noreply@anthropic.com>

.github/workflows/validate.yml

@@ -718,6 +718,9 @@ jobs:
                 --exclude="install-codex.sh" \
                 --exclude="install-codex-plugin.sh" \
                 --exclude="install-codex-native-skills.sh" \
+                --exclude="install-claude.sh" \
+                --exclude="install-agy.sh" \
+                --exclude="install.sh" \
                 --exclude="ci-local-release.sh" \
                 . 2>/dev/null; then
               echo "Found: $pattern"

docs/contracts/context-map.md

@@ -103,10 +103,8 @@ and [CDLC](https://github.com/boshu2/agentops/blob/main/docs/cdlc.md) for the ar
 
 ```mermaid
 graph LR
-  acfs -- "supplier-to" --> operating-loop-skill
   agy-headless-evidence -- "customer-of" --> agy-native
   agy-headless-evidence -- "supplier-to" --> validate
-  agy-native -- "customer-of" --> operating-loop-skill
   automation-shape-routing -- "supplier-to" --> operationalize
   automation-shape-routing -- "supplier-to" --> skill-builder
   automation-shape-routing -- "supplier-to" --> workflow-builder
@@ -175,7 +173,6 @@ graph LR
 | `agent-native` | produces | docs/contracts/agent-runtime-profile.md |
 | `agy-headless-evidence` | consumes | agy-native |
 | `agy-headless-evidence` | produces | agy-evidence-dir |
-| `agy-native` | consumes | operating-loop-skill |
 | `agy-native` | produces | agy-run-evidence |
 | `autodev` | consumes | evolve |
 | `autodev` | consumes | rpi |
@@ -194,7 +191,7 @@ graph LR
 | `continuity-loop` | consumes | ntm |
 | `continuity-loop` | produces | .agents/continuity/state.json |
 | `continuity-loop` | produces | escalation-message |
-| `converge` | consumes | ao converge |
+| `converge` | consumes | command-help |
 | `converge` | produces | stdout |
 | `converter` | produces | converted-skill |
 | `council` | consumes | standards |
@@ -208,8 +205,6 @@ graph LR
 | `crank` | produces | .agents/swarm/results/*.json |
 | `crank` | produces | git-changes |
 | `curate` | produces | .agents/research/*.md |
-| `discovery` | consumes | brainstorm |
-| `discovery` | consumes | design |
 | `discovery` | consumes | plan |
 | `discovery` | consumes | pre-mortem |
 | `discovery` | consumes | research |
@@ -221,7 +216,6 @@ graph LR
 | `doc` | produces | documentation |
 | `domain` | produces | stdout |
 | `eval-outcomes` | consumes | council |
-| `eval-outcomes` | consumes | ratchet |
 | `eval-outcomes` | consumes | validate |
 | `eval-outcomes` | produces | skills/council/schemas/verdict.json |
 | `evolve` | consumes | compile |
@@ -237,8 +231,6 @@ graph LR
 | `implement` | consumes | domain |
 | `implement` | produces | git-changes |
 | `operationalize` | consumes | .agents/research/*.md |
-| `operationalize` | consumes | audit-report |
-| `operationalize` | consumes | learning |
 | `operationalize` | produces | .agents/operationalize/*.md |
 | `operationalize` | produces | routed-handoffs |
 | `perf` | consumes | repo-context |
@@ -268,7 +260,6 @@ graph LR
 | `recover` | produces | .agents/rpi/*.md |
 | `red-team` | consumes | repo-context |
 | `red-team` | produces | result.json |
-| `refactor` | consumes | complexity |
 | `refactor` | consumes | repo-context |
 | `refactor` | produces | git-changes |
 | `release` | produces | result.json |
@@ -282,7 +273,6 @@ graph LR
 | `rpi` | consumes | crank |
 | `rpi` | consumes | discovery |
 | `rpi` | consumes | domain |
-| `rpi` | consumes | ratchet |
 | `rpi` | consumes | validate |
 | `rpi` | produces | .agents/rpi/*.md |
 | `scaffold` | produces | converted-skill |

docs/plans/2026-06-12-dirty-main-attribution.md

@@ -13,9 +13,9 @@ bead: ag-xwjlc (seams epic)
 - main's recent history is `aug/*` merge commits (aug/validate, aug/status,
   aug/rpi, aug/review, aug/refactor ...) — an augmentation wave in progress.
 - The working tree on host main is dirty with a large in-flight pass:
-  PRODUCT.md, docs/SKILLS.md, docs/contracts/* (context-map, dispositions,
+  PRODUCT.md, docs/SKILLS.md, `docs/contracts/*` (context-map, dispositions,
   critical-skills), .github/workflows/validate.yml, skills/SKILL-TIERS.md, many
-  skills/*/SKILL.md (including uncommitted DELETIONS of vibe/ratchet/brainstorm/
+  `skills/*/SKILL.md` (including uncommitted DELETIONS of vibe/ratchet/brainstorm/
   design/scenario/... — at committed HEAD those skills still exist), full
   skills-codex/ + images/gemini/ regen surfaces. `git status --porcelain`
   snapshot captured at decision time (~600 entries; see session evidence

docs/plans/ag-s43tg-resume-and-mto-alignment.md

@@ -63,8 +63,8 @@ hard constraint reserves for one writer:
    scenario-test linkage, regen-all --check, context-map drift). Fix-forward on
    red; the gate naming a stale surface is the designed trip-wire, not a thing
    to disarm.
-6. **One atomic commit** (removals + ledger + regen + the two tracked plan docs
-   + trigger manifest), push main (carries 3384f5638 with it).
+6. **One atomic commit** (removals + ledger + regen + the two tracked plan
+   docs + trigger manifest), push main (carries 3384f5638 with it).
    `git revert <sha>` restores everything together.
 7. **Verify:** `ls skills | wc -l` = 68; pinned-manifest grep all-GREEN;
    `link-skill --all --relink`; `validate.yml` green on main.

evidence/skill-prune-phase2.md

@@ -27,8 +27,8 @@ Skill count: 114 → **105** (97 user-facing + 8 internal).
 
 - `scripts/skill-flow-standalone.txt`: 3 stale entries removed.
 - `scripts/validate-agy-plugin.sh`: 10 stale core_skills removed (9 were Phase 1
-  misses — script was failing on main); `.agy-plugin/skills/` bundle pruned to 17
-  + 9 drifted SKILL.mds resynced. Validator now PASSES (was failing).
+  misses — script was failing on main); `.agy-plugin/skills/` bundle pruned to
+  17 + 9 drifted SKILL.mds resynced. Validator now PASSES (was failing).
 - `tests/docs/broken-links-allowlist.txt`: matches belong to the research skill's
   own reference files (same basename) — no change needed.
 - zsh footnote: the first sweep pass used unquoted `$names` word-splitting, which

registry.json

@@ -1,6 +1,6 @@
 {
   "schema_version": 2,
-  "generated_at": "2026-06-14T17:07:44Z",
+  "generated_at": "2026-06-14T18:25:10Z",
   "summary": {
     "skills": 72,
     "hooks": 0,
@@ -1749,9 +1749,7 @@
       "parity_policy": "required",
       "aliases": [],
       "supersedes": null,
-      "consumes": [
-        "operating-loop-skill"
-      ],
+      "consumes": [],
       "produces": [
         "agy-run-evidence"
       ],
@@ -2163,7 +2161,7 @@
       "aliases": [],
       "supersedes": null,
       "consumes": [
-        "ao converge"
+        "command-help"
       ],
       "produces": [
         "stdout"
@@ -2362,8 +2360,6 @@
       "aliases": [],
       "supersedes": null,
       "consumes": [
-        "brainstorm",
-        "design",
         "plan",
         "pre-mortem",
         "research",
@@ -2472,7 +2468,6 @@
       "supersedes": null,
       "consumes": [
         "validate",
-        "ratchet",
         "council"
       ],
       "produces": [
@@ -2827,9 +2822,7 @@
       "aliases": [],
       "supersedes": null,
       "consumes": [
-        ".agents/research/*.md",
-        "audit-report",
-        "learning"
+        ".agents/research/*.md"
       ],
       "produces": [
         ".agents/operationalize/*.md",
@@ -3253,7 +3246,6 @@
       "aliases": [],
       "supersedes": null,
       "consumes": [
-        "complexity",
         "repo-context"
       ],
       "produces": [
@@ -3387,7 +3379,6 @@
         "crank",
         "discovery",
         "domain",
-        "ratchet",
         "validate"
       ],
       "produces": [

scripts/skill-flow-standalone.txt

@@ -75,3 +75,5 @@ stash-hygiene-sweep              # standalone external-boundary or library/analy
 ubs                              # standalone external-boundary or library/analyzer skill; no peer skill edge by design
 vibing-with-ntm                  # standalone external-boundary or library/analyzer skill; no peer skill edge by design
 worktree-branch-rationalization  # standalone external-boundary or library/analyzer skill; no peer skill edge by design
+beads-bv                         # tracker-triage primitive; lost its only incoming edge when operating-loop-skill was pruned (ag-if7p) — leaf by design, like status/beads-workflow
+converge                         # thin memo over the `ao converge` command (driving-adapter); no peer skill edge by design, like status

skills-codex/.agentops-manifest.json

@@ -1000,7 +1000,7 @@
     {
       "name": "acfs",
       "source_skill": "skills/acfs",
-      "source_hash": "b0e2e35f958f5fa111dccc0ea1971d71c065a98303821c9d80a5c1412fc51faa",
+      "source_hash": "a0ec97af82529e00df46d325121b2e45621f8607dac2c3545364b87e91676979",
       "generated_hash": "bf2564f612c7edeeb8a96b72649a33a2d8247b3b30a9ae52603d02c060bfb946"
     },
     {
@@ -1030,7 +1030,7 @@
     {
       "name": "agy-native",
       "source_skill": "skills/agy-native",
-      "source_hash": "85309d247ad37bd783f62c3e554af93a377de80e07bfd2674dd26bcd824a58d0",
+      "source_hash": "b84c5f7d13b5665ebf8f3b7aa6e3408f20353022e9f05a1f1122316e39a0035a",
       "generated_hash": "38f0627619fe29459b27ae125364737483302cdf18a77e13d6f4e3aef21ea34d"
     },
     {
@@ -1144,7 +1144,7 @@
     {
       "name": "converge",
       "source_skill": "skills/converge",
-      "source_hash": "f7711e6ff9a7d4c212cd8a9242c3189649290bf96a2ba50074e16f9a9d8ce094",
+      "source_hash": "e73d0e90e5199e9b5c38bccfe11c10773ff1da131c6bf9f3cb8f673f836128ba",
       "generated_hash": "4210749b835a469aac459ae9a6e66aaab9df62e2370f02575052f6eaf97b6264"
     },
     {
@@ -1210,7 +1210,7 @@
     {
       "name": "discovery",
       "source_skill": "skills/discovery",
-      "source_hash": "f41d6f1c8d863cc8b5f848286a3e15a9f613158b0671bbe9a37c0d2001563288",
+      "source_hash": "b335f6eca9b88cf0878a788002052d1d3a926fcde2a58689786213eaabbe9f27",
       "generated_hash": "38ab1cbc2a714653d6d0a3c675d8336bc3f6e4dfe6f388016573a5d4f26f9491"
     },
     {
@@ -1246,7 +1246,7 @@
     {
       "name": "eval-outcomes",
       "source_skill": "skills/eval-outcomes",
-      "source_hash": "7f05f9bc5d0220e07ba01a51c90fc2fe9c552b885373e5db769019dde0cd37e1",
+      "source_hash": "064d6cc2cbc4112af98f4c5798e50c734974e76394b38e86f6d601b0e66bb110",
       "generated_hash": "a7243ecfd6177318f1ef25babdeb0b9061d09b9c2f8d89495268ed1c46925bd0"
     },
     {
@@ -1354,7 +1354,7 @@
     {
       "name": "operationalize",
       "source_skill": "skills/operationalize",
-      "source_hash": "c0e8a3458553325e204a7ea5793b6066257eaa8fe1be105a5de2bd590ae1afcf",
+      "source_hash": "82bcecb5ebf831dd49cdafda963293d84bfd9ce8d5ff224b9ba8f588e51d3b5a",
       "generated_hash": "c2916dde8569046457c2b338df46f83edca1d8211892ff9acc2cd896920e807c"
     },
     {
@@ -1366,7 +1366,7 @@
     {
       "name": "perf",
       "source_skill": "skills/perf",
-      "source_hash": "3af782988417a0d8ee32ef06801c3b8eb5a4e6d1c7c22a06369ff0e9b807db57",
+      "source_hash": "09389093bec9cd998d75ce784378f23e8cc10d58a7186fd8a349d70300336496",
       "generated_hash": "044755509c7a50a2b6d10fc598b3197f826767220bffc5958b1cf3a7580885b2"
     },
     {
@@ -1492,7 +1492,7 @@
     {
       "name": "refactor",
       "source_skill": "skills/refactor",
-      "source_hash": "5a0a4e7eea7df6453c444902c54bfeccb8a907e2cf71e3cb44f74392d9cfcaa6",
+      "source_hash": "8fecbf97ac26d84ebd29c78360c8aa940139aeb60deedfb776e183a9c1cfff39",
       "generated_hash": "fa262113610755ebedd60a0c21d1a7f0b8359f215d05f67abd26360a03faeb0c"
     },
     {
@@ -1540,7 +1540,7 @@
     {
       "name": "rpi",
       "source_skill": "skills/rpi",
-      "source_hash": "b87afe04860b7b3926f3647a99a05fcc22cbb6697f59ab4bd4a2a1a58943bafb",
+      "source_hash": "20fa14f17983163bdf55e626b18e2146ed4571fe00c8fba7eebfdd6587c67ab5",
       "generated_hash": "9e1a058756851e44572b3992f93aa0377b60cdb8bff67c2ebb0310fb5db6f9bc"
     },
     {
@@ -1666,7 +1666,7 @@
     {
       "name": "test",
       "source_skill": "skills/test",
-      "source_hash": "388e3bc02cf172fa69fdeab61b044d8aaca5f8d79d9ef679e9e6926a5f13004e",
+      "source_hash": "700d7605b340dd7ead3bb7169a7ea4d51ba08235c3372ad226d06b9e4cbda6b7",
       "generated_hash": "82bfb16f9fbac7fce26a44c7cac7fd512fc4bb28464eda5adb7e2bd72b09a6be"
     },
     {

skills-codex/acfs/.agentops-generated.json

@@ -2,6 +2,6 @@
   "generator": "manual-maintained",
   "source_skill": "skills/acfs",
   "layout": "modular",
-  "source_hash": "b0e2e35f958f5fa111dccc0ea1971d71c065a98303821c9d80a5c1412fc51faa",
+  "source_hash": "a0ec97af82529e00df46d325121b2e45621f8607dac2c3545364b87e91676979",
   "generated_hash": "bf2564f612c7edeeb8a96b72649a33a2d8247b3b30a9ae52603d02c060bfb946"
 }

skills-codex/agy-native/.agentops-generated.json

@@ -2,6 +2,6 @@
   "generator": "manual-maintained",
   "source_skill": "skills/agy-native",
   "layout": "modular",
-  "source_hash": "85309d247ad37bd783f62c3e554af93a377de80e07bfd2674dd26bcd824a58d0",
+  "source_hash": "b84c5f7d13b5665ebf8f3b7aa6e3408f20353022e9f05a1f1122316e39a0035a",
   "generated_hash": "38f0627619fe29459b27ae125364737483302cdf18a77e13d6f4e3aef21ea34d"
 }

skills-codex/converge/.agentops-generated.json

@@ -2,6 +2,6 @@
   "generator": "manual-maintained",
   "source_skill": "skills/converge",
   "layout": "modular",
-  "source_hash": "f7711e6ff9a7d4c212cd8a9242c3189649290bf96a2ba50074e16f9a9d8ce094",
+  "source_hash": "e73d0e90e5199e9b5c38bccfe11c10773ff1da131c6bf9f3cb8f673f836128ba",
   "generated_hash": "4210749b835a469aac459ae9a6e66aaab9df62e2370f02575052f6eaf97b6264"
 }

skills-codex/discovery/.agentops-generated.json

@@ -2,6 +2,6 @@
   "generator": "manual-maintained",
   "source_skill": "skills/discovery",
   "layout": "modular",
-  "source_hash": "f41d6f1c8d863cc8b5f848286a3e15a9f613158b0671bbe9a37c0d2001563288",
+  "source_hash": "b335f6eca9b88cf0878a788002052d1d3a926fcde2a58689786213eaabbe9f27",
   "generated_hash": "38ab1cbc2a714653d6d0a3c675d8336bc3f6e4dfe6f388016573a5d4f26f9491"
 }

skills-codex/eval-outcomes/.agentops-generated.json

@@ -2,6 +2,6 @@
   "generator": "manual-maintained",
   "source_skill": "skills/eval-outcomes",
   "layout": "stub",
-  "source_hash": "7f05f9bc5d0220e07ba01a51c90fc2fe9c552b885373e5db769019dde0cd37e1",
+  "source_hash": "064d6cc2cbc4112af98f4c5798e50c734974e76394b38e86f6d601b0e66bb110",
   "generated_hash": "a7243ecfd6177318f1ef25babdeb0b9061d09b9c2f8d89495268ed1c46925bd0"
 }

skills-codex/operationalize/.agentops-generated.json

@@ -2,6 +2,6 @@
   "generator": "manual-maintained",
   "source_skill": "skills/operationalize",
   "layout": "modular",
-  "source_hash": "c0e8a3458553325e204a7ea5793b6066257eaa8fe1be105a5de2bd590ae1afcf",
+  "source_hash": "82bcecb5ebf831dd49cdafda963293d84bfd9ce8d5ff224b9ba8f588e51d3b5a",
   "generated_hash": "c2916dde8569046457c2b338df46f83edca1d8211892ff9acc2cd896920e807c"
 }

skills-codex/perf/.agentops-generated.json

@@ -2,6 +2,6 @@
   "generator": "manual-maintained",
   "source_skill": "skills/perf",
   "layout": "modular",
-  "source_hash": "3af782988417a0d8ee32ef06801c3b8eb5a4e6d1c7c22a06369ff0e9b807db57",
+  "source_hash": "09389093bec9cd998d75ce784378f23e8cc10d58a7186fd8a349d70300336496",
   "generated_hash": "044755509c7a50a2b6d10fc598b3197f826767220bffc5958b1cf3a7580885b2"
 }

skills-codex/refactor/.agentops-generated.json

@@ -2,6 +2,6 @@
   "generator": "manual-maintained",
   "source_skill": "skills/refactor",
   "layout": "modular",
-  "source_hash": "5a0a4e7eea7df6453c444902c54bfeccb8a907e2cf71e3cb44f74392d9cfcaa6",
+  "source_hash": "8fecbf97ac26d84ebd29c78360c8aa940139aeb60deedfb776e183a9c1cfff39",
   "generated_hash": "fa262113610755ebedd60a0c21d1a7f0b8359f215d05f67abd26360a03faeb0c"
 }

skills-codex/rpi/.agentops-generated.json

@@ -2,6 +2,6 @@
   "generator": "manual-maintained",
   "source_skill": "skills/rpi",
   "layout": "modular",
-  "source_hash": "b87afe04860b7b3926f3647a99a05fcc22cbb6697f59ab4bd4a2a1a58943bafb",
+  "source_hash": "20fa14f17983163bdf55e626b18e2146ed4571fe00c8fba7eebfdd6587c67ab5",
   "generated_hash": "9e1a058756851e44572b3992f93aa0377b60cdb8bff67c2ebb0310fb5db6f9bc"
 }

skills-codex/test/.agentops-generated.json

@@ -2,6 +2,6 @@
   "generator": "manual-maintained",
   "source_skill": "skills/test",
   "layout": "modular",
-  "source_hash": "388e3bc02cf172fa69fdeab61b044d8aaca5f8d79d9ef679e9e6926a5f13004e",
+  "source_hash": "700d7605b340dd7ead3bb7169a7ea4d51ba08235c3372ad226d06b9e4cbda6b7",
   "generated_hash": "82bfb16f9fbac7fce26a44c7cac7fd512fc4bb28464eda5adb7e2bd72b09a6be"
 }

skills/acfs/SKILL.md

@@ -11,9 +11,7 @@ hexagonal_role: driving-adapter
 consumes: []
 produces:
 - substrate-health-report
-context_rel:
-- kind: supplier-to
-  with: operating-loop-skill
+context_rel: []
 skill_api_version: 1
 user-invocable: true
 context:
@@ -25,7 +23,7 @@ context:
   intel_scope: topic
 metadata:
   tier: orchestration
-  dependencies: [ntm, account-rotation, cass-memory, dcg]
+  dependencies: [ntm, account-rotation, dcg]
   external_dependencies: [fleet-ops]
   stability: stable
 output_contract: "Terminal report (acfs doctor/init verdict) + a stated next action; no files written unless a sub-tool persists state (br/cm)."

skills/agy-native/SKILL.md

@@ -7,13 +7,10 @@ practices:
 - team-topologies
 - continuous-delivery
 hexagonal_role: driving-adapter
-consumes:
-- operating-loop-skill
+consumes: []
 produces:
 - agy-run-evidence
-context_rel:
-- kind: customer-of
-  with: operating-loop-skill
+context_rel: []
 skill_api_version: 1
 user-invocable: false
 context:

skills/catalog.json

@@ -1,6 +1,6 @@
 {
   "schema_version": "1",
-  "generated_at": "2026-06-13T18:04:08Z",
+  "generated_at": "2026-06-14T18:26:50Z",
   "skill_count": 72,
   "skills": [
     {
@@ -30,12 +30,7 @@
         "pragmatic-programmer",
         "continuous-delivery"
       ],
-      "context_rel": [
-        {
-          "kind": "supplier-to",
-          "with": "operating-loop-skill"
-        }
-      ],
+      "context_rel": [],
       "references_count": 0,
       "codex_override_present": true
     },
@@ -107,9 +102,7 @@
       "description": "|-",
       "hexagonal_role": "driving-adapter",
       "user_invocable": false,
-      "consumes": [
-        "operating-loop-skill"
-      ],
+      "consumes": [],
       "produces": [
         "agy-run-evidence"
       ],
@@ -117,12 +110,7 @@
         "team-topologies",
         "continuous-delivery"
       ],
-      "context_rel": [
-        {
-          "kind": "customer-of",
-          "with": "operating-loop-skill"
-        }
-      ],
+      "context_rel": [],
       "references_count": 1,
       "codex_override_present": true
     },
@@ -410,7 +398,7 @@
       "hexagonal_role": "driving-adapter",
       "user_invocable": false,
       "consumes": [
-        "ao converge"
+        "command-help"
       ],
       "produces": [
         "stdout"
@@ -421,12 +409,12 @@
       ],
       "context_rel": [],
       "references_count": 0,
-      "codex_override_present": false
+      "codex_override_present": true
     },
     {
       "name": "converter",
       "description": "Convert AgentOps skill formats.",
-      "hexagonal_role": "generic",
+      "hexagonal_role": "driven-adapter",
       "user_invocable": false,
       "consumes": [],
       "produces": [
@@ -533,8 +521,6 @@
       "hexagonal_role": "domain",
       "user_invocable": true,
       "consumes": [
-        "brainstorm",
-        "design",
         "plan",
         "pre-mortem",
         "research",
@@ -604,7 +590,6 @@
       "user_invocable": false,
       "consumes": [
         "validate",
-        "ratchet",
         "council"
       ],
       "produces": [
@@ -621,7 +606,7 @@
     {
       "name": "evolve",
       "description": "Run autonomous improvement loops.",
-      "hexagonal_role": "supporting",
+      "hexagonal_role": "domain",
       "user_invocable": true,
       "consumes": [
         "rpi",
@@ -813,9 +798,7 @@
       "hexagonal_role": "domain",
       "user_invocable": false,
       "consumes": [
-        ".agents/research/*.md",
-        "audit-report",
-        "learning"
+        ".agents/research/*.md"
       ],
       "produces": [
         ".agents/operationalize/*.md",
@@ -964,7 +947,7 @@
     },
     {
       "name": "pre-land-refuters",
-      "description": "Dispatch unbiased refuters (fresh Fable + read-only codex exec) to attack a completion claim before landing a large change. Triggers: pre-land validation, refute before push.",
+      "description": "Dispatch fresh-context refuters (model-agnostic by default; multi-model opt-in) to attack a completion claim at the mutate-shared-trunk pawl before landing, at any complexity. Triggers: pre-land validation, refute pre-push.",
       "hexagonal_role": "driving-adapter",
       "user_invocable": true,
       "consumes": [
@@ -1150,7 +1133,6 @@
       "hexagonal_role": "supporting",
       "user_invocable": false,
       "consumes": [
-        "complexity",
         "repo-context"
       ],
       "produces": [
@@ -1239,13 +1221,12 @@
     {
       "name": "rpi",
       "description": "Run discovery, crank, validation.",
-      "hexagonal_role": "supporting",
+      "hexagonal_role": "domain",
       "user_invocable": true,
       "consumes": [
         "crank",
         "discovery",
         "domain",
-        "ratchet",
         "validate"
       ],
       "produces": [

skills/converge/SKILL.md

@@ -6,7 +6,7 @@ practices:
 - llm-eval-harness
 hexagonal_role: driving-adapter
 consumes:
-- ao converge
+- command-help
 produces:
 - stdout
 context_rel: []

skills/discovery/SKILL.md

@@ -7,8 +7,6 @@ practices:
 - mythical-man-month
 hexagonal_role: domain
 consumes:
-- brainstorm
-- design
 - plan
 - pre-mortem
 - research
@@ -33,8 +31,6 @@ context:
 metadata:
   tier: meta
   dependencies:
-  - brainstorm
-  - design
   - research
   - plan
   - pre-mortem

skills/eval-outcomes/SKILL.md

@@ -8,7 +8,6 @@ practices:
 hexagonal_role: supporting
 consumes:
 - validate
-- ratchet
 - council
 produces:
 - skills/council/schemas/verdict.json
@@ -21,7 +20,7 @@ context:
   intel_scope: topic
 metadata:
   tier: execution
-  dependencies: [validate, ratchet]
+  dependencies: [validate]
   stability: experimental
 output_contract: "skills/council/schemas/verdict.json (one council verdict record)"
 ---

skills/operationalize/SKILL.md

@@ -9,8 +9,6 @@ practices:
 hexagonal_role: domain
 consumes:
 - .agents/research/*.md
-- audit-report
-- learning
 produces:
 - .agents/operationalize/*.md
 - routed-handoffs

skills/perf/SKILL.md

@@ -26,7 +26,6 @@ metadata:
   tier: execution
   dependencies:
   - standards
-  - complexity
 output_contract: 'stdout: performance profile/benchmark report'
 ---
 # Perf Skill

skills/refactor/SKILL.md

@@ -7,7 +7,6 @@ practices:
 - design-patterns
 hexagonal_role: supporting
 consumes:
-- complexity
 - repo-context
 produces:
 - git-changes
@@ -25,7 +24,6 @@ metadata:
   tier: execution
   dependencies:
   - standards
-  - complexity
   - beads
 output_contract: code changes with regression verification
 ---

skills/rpi/SKILL.md

@@ -15,7 +15,6 @@ consumes:
 - crank
 - discovery
 - domain
-- ratchet
 - validate
 produces:
 - .agents/rpi/*.md
@@ -43,7 +42,6 @@ metadata:
   - discovery
   - crank
   - validate
-  - ratchet
   internal: false
 output_contract: .agents/rpi/YYYY-MM-DD-*.md
 ---

skills/test/SKILL.md

@@ -25,7 +25,6 @@ metadata:
   tier: execution
   dependencies:
   - standards
-  - complexity
 output_contract: test files, coverage report
 ---
 # Test Skill

tests/scripts/brainstorm-discovery-ideation.bats

@@ -89,22 +89,24 @@ assert_lacks() {
     done
 }
 
-@test "bead-operationalization reference uses bd and bans br/bv tracker refs" {
-    assert_has "$REF_BEADS" "bd create"
-    assert_has "$REF_BEADS" "bd dep add"
+@test "bead-operationalization reference uses br and bans the retired bd tracker" {
+    # Tracker contracts migrated bd -> br (commit f650d41bb); the reference docs
+    # are now repo-native br. The test follows the migration: assert br, ban bd.
+    assert_has "$REF_BEADS" "br create"
+    assert_has "$REF_BEADS" "br dep add"
     assert_has "$REF_BEADS" "DO NOT OVERSIMPLIFY"
     assert_has "$REF_BEADS" "DO NOT LOSE FEATURES"
-    # idea-wizard's br/bv tracker must NOT leak into the ported repo-native docs.
-    assert_lacks_regex "$REF_BEADS"    '\b(br|bv)\s+(list|create|dep|ready|--robot)'
-    assert_lacks_regex "$REF_IDEATION" '\b(br|bv)\s+(list|create|dep|ready|--robot)'
-    assert_lacks_regex "$REF_RUBRIC"   '\b(br|bv)\s+(list|create|dep|ready|--robot)'
+    # The retired bd tracker must NOT leak back into the repo-native docs.
+    assert_lacks_regex "$REF_BEADS"    '\bbd\s+(list|create|dep|ready|--robot)'
+    assert_lacks_regex "$REF_IDEATION" '\bbd\s+(list|create|dep|ready|--robot)'
+    assert_lacks_regex "$REF_RUBRIC"   '\bbd\s+(list|create|dep|ready|--robot)'
 }
 
 @test "ideation-mode reference documents the mode-selection rule and grounding" {
     assert_has "$REF_IDEATION" "When to use which mode"
     assert_has "$REF_IDEATION" "AGENTS.md"
-    assert_has "$REF_IDEATION" "bd list --json"
-    assert_has "$REF_IDEATION" "bd list --status closed --json"
+    assert_has "$REF_IDEATION" "br list --json"
+    assert_has "$REF_IDEATION" "br list --status closed --json"
 }
 
 @test "discovery SKILL wires the open-ended ideate path with operationalize and refine" {
@@ -130,11 +132,13 @@ assert_lacks() {
     assert_has "$CODEX_BRAINSTORM" '--ideate'
     assert_has "$CODEX_DISCOVERY" "Open-Ended Path"
     assert_has "$CODEX_DISCOVERY" 'internal modes (absorbed, ag-s43tg)'
-    # Codex bodies must not leak Claude-era primitives (parity rule).
-    assert_lacks "$CODEX_BRAINSTORM" "AskUserQuestion"
+    # Codex bodies must not leak Claude-era primitives (parity rule). The
+    # brainstorm reference (goal-clarification-brainstorm.md) is the CLAUDE-side
+    # surface and legitimately uses AskUserQuestion, so the anti-leak guard
+    # applies to the genuine codex twin (skills-codex/discovery/SKILL.md), not
+    # to the shared Claude reference.
     assert_lacks "$CODEX_DISCOVERY" "AskUserQuestion"
     # Codex must use $skill notation, not /skill, for the new content.
-    assert_lacks "$CODEX_BRAINSTORM" "claude -p"
     assert_lacks "$CODEX_DISCOVERY" "/brainstorm --ideate"
 }
 

tests/scripts/install-sh-runtime-detection.bats

@@ -74,7 +74,10 @@ run_install_with_path() {
   [[ "$output" == *"Detected Claude Code: yes"* ]]
   [[ "$output" == *"Detected Codex CLI: no"* ]]
   [[ "$output" == *"single-runtime mode (Claude Code only)"* ]]
-  [[ "$output" == *"install Codex CLI: https://github.com/openai/codex"* ]]
+  # install.sh now points single-runtime users at the other runtimes' install
+  # links (multi-vendor: Codex + Gemini/AGY) under a "install another runtime"
+  # banner. Assert the Codex link is offered.
+  [[ "$output" == *"Codex CLI:   https://github.com/openai/codex"* ]]
 }
 
 @test "codex only: single-runtime + Claude install suggestion" {
@@ -88,7 +91,8 @@ run_install_with_path() {
   [[ "$output" == *"Detected Claude Code: no"* ]]
   [[ "$output" == *"Detected Codex CLI: yes"* ]]
   [[ "$output" == *"single-runtime mode (Codex CLI only)"* ]]
-  [[ "$output" == *"install Claude Code: https://docs.anthropic.com/en/docs/claude-code"* ]]
+  # Assert the Claude install link is offered to the single-runtime Codex user.
+  [[ "$output" == *"Claude Code: https://docs.anthropic.com/en/docs/claude-code"* ]]
 }
 
 @test "neither present: warning + both install links" {