@cleocode/skills 2026.10.27 → 2026.11.1-canary.11

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (47) hide show
  1. package/README.md +3 -2
  2. package/package.json +1 -1
  3. package/profiles/recommended.json +0 -1
  4. package/skills/_shared/__tests__/loom-adr-links.test.ts +5 -1
  5. package/skills/_shared/__tests__/loom-stage-coverage.test.ts +14 -5
  6. package/skills/_shared/subagent-protocol-base.cant +7 -7
  7. package/skills/_shared/subagent-protocol-base.md +10 -6
  8. package/skills/ct-adr-recorder/SKILL.md +3 -3
  9. package/skills/ct-cleo/SKILL.md +6 -6
  10. package/skills/ct-cleo/references/memory.md +6 -0
  11. package/skills/ct-cleo/references/planning-loop.md +7 -5
  12. package/skills/ct-cleo/references/session-protocol.md +6 -8
  13. package/skills/ct-codebase-mapper/SKILL.md +27 -3
  14. package/skills/ct-consensus-voter/SKILL.md +3 -3
  15. package/skills/ct-contribution/SKILL.md +4 -4
  16. package/skills/ct-dev-workflow/SKILL.md +130 -86
  17. package/skills/ct-documentor/SKILL.md +73 -9
  18. package/skills/ct-epic-architect/SKILL.md +4 -4
  19. package/skills/ct-grade/SKILL.md +3 -3
  20. package/skills/ct-grade/references/domains.md +1 -1
  21. package/skills/ct-grade/references/grade-spec.md +1 -1
  22. package/skills/ct-grade/references/scenario-playbook.md +1 -1
  23. package/skills/ct-grade/references/token-tracking.md +1 -1
  24. package/skills/ct-lead/SKILL.md +37 -19
  25. package/skills/ct-lead/references/aggregation-protocol.md +3 -2
  26. package/skills/ct-lead/references/spawn-pattern.md +14 -58
  27. package/skills/ct-orchestrator/SKILL.md +244 -38
  28. package/skills/ct-orchestrator/references/SUBAGENT-PROTOCOL-BLOCK.md +5 -5
  29. package/skills/ct-orchestrator/references/orchestrator-compliance.md +5 -5
  30. package/skills/ct-orchestrator/references/orchestrator-patterns.md +20 -22
  31. package/skills/ct-orchestrator/references/orchestrator-tokens.md +1 -1
  32. package/skills/ct-release-orchestrator/SKILL.md +58 -14
  33. package/skills/ct-research-agent/SKILL.md +4 -4
  34. package/skills/ct-spec-writer/SKILL.md +4 -4
  35. package/skills/ct-task-executor/SKILL.md +28 -13
  36. package/skills/ct-task-executor/references/acceptance-criteria-mapping.md +1 -1
  37. package/skills/ct-task-executor/references/anti-patterns.md +4 -4
  38. package/skills/ct-task-executor/references/evidence-and-gates.md +43 -1
  39. package/skills/ct-task-executor/references/implementation-patterns.md +6 -5
  40. package/skills/ct-validator/SKILL.md +14 -20
  41. package/skills/ct-validator/references/compliance-reports.md +1 -16
  42. package/skills/manifest.json +27 -107
  43. package/skills/ct-ivt-looper/SKILL.md +0 -220
  44. package/skills/ct-ivt-looper/manifest-entry.json +0 -30
  45. package/skills/ct-ivt-looper/references/escalation.md +0 -91
  46. package/skills/ct-ivt-looper/references/frameworks.md +0 -119
  47. package/skills/ct-ivt-looper/references/loop-anatomy.md +0 -156
package/README.md CHANGED
@@ -1,5 +1,7 @@
1
1
  # @cleocode/skills
2
2
 
3
+ > Part of [CLEO](https://github.com/kryptobaseddev/cleo#readme). Start with the root README; run `cleo <command> --help` for any command.
4
+
3
5
  CLEO skill definitions - bundled capabilities for AI agents.
4
6
 
5
7
  ## Overview
@@ -59,9 +61,8 @@ installability and every documented `cleo` command honest.
59
61
  | **ct-contribution** | Guided workflow for multi-agent consensus contributions. |
60
62
  | **ct-council** | Convene "The Council" — a 5-advisor, shuffled gate-based peer-review, chairman-synthesis workflow for reviewin… |
61
63
  | **ct-epic-architect** | Epic planning and task decomposition for breaking down large initiatives into atomic, executable tasks. |
62
- | **ct-ivt-looper** | Runs a project-agnostic autonomous Implement-then-Validate-then-Test compliance loop on any git worktree. |
63
64
  | **ct-provenance-keeper** | Generates in-toto v1 attestations, SLSA-level provenance records, SBOMs (CycloneDX or SPDX), and sigstore/cosi… |
64
- | **ct-release-orchestrator** | Orchestrates the canonical 4-verb release pipeline introduced by SPEC-T9345: cleo release plan, then cleo rele… |
65
+ | **ct-release-orchestrator** | Orchestrates CLEO releases: `cleo release cut <version>` tags the newest main commit with green CI, and release.yml writes the CHANGELOG from changesets and publishes. |
65
66
  | **ct-research-agent** | Multi-source research and investigation combining web search, documentation lookup via Context7, and codebase … |
66
67
  | **ct-spec-writer** | Technical specification writing using RFC 2119 language for clear, unambiguous requirements. |
67
68
  | **ct-validator** | Compliance validation for verifying systems, documents, or code against requirements, schemas, or standards. |
package/package.json CHANGED
@@ -1,6 +1,6 @@
1
1
  {
2
2
  "name": "@cleocode/skills",
3
- "version": "2026.10.27",
3
+ "version": "2026.11.1-canary.11",
4
4
  "description": "CLEO skill definitions - bundled with CLEO monorepo",
5
5
  "main": "index.js",
6
6
  "types": "index.d.ts",
@@ -9,7 +9,6 @@
9
9
  "ct-spec-writer",
10
10
  "ct-epic-architect",
11
11
  "ct-validator",
12
- "ct-ivt-looper",
13
12
  "ct-release-orchestrator",
14
13
  "ct-contribution",
15
14
  "ct-artifact-publisher",
@@ -8,7 +8,7 @@
8
8
  * 2. Every ADR id in the array resolves to a real file under `.cleo/adrs/`.
9
9
  *
10
10
  * The mapping of (stage -> required adrRefs minimum) is the authoritative
11
- * source-of-truth defined in `docs/skills/loom-coverage-matrix.md` under
11
+ * source-of-truth defined in `docs/reference/loom-coverage-matrix.md` under
12
12
  * the "ADR Bindings Section". When that doc updates, this test updates.
13
13
  *
14
14
  * @task T9665
@@ -96,6 +96,10 @@ for (const s of manifest.skills) {
96
96
  skillsByStage.set(s.loomStage, s);
97
97
  }
98
98
  }
99
+ // T14311: `testing` shares the validation skill (ct-validator) since
100
+ // ct-ivt-looper was removed, so it is held to the same ADR floor through it.
101
+ const validationSkill = skillsByStage.get('validation');
102
+ if (validationSkill) skillsByStage.set('testing', validationSkill);
99
103
 
100
104
  // ---------------------------------------------------------------------------
101
105
  // Gate 1: every LOOM-stage skill declares a non-empty adrRefs[]
@@ -82,6 +82,12 @@ const CROSS_CUTTING_PROTOCOLS = new Set([
82
82
  /**
83
83
  * Skill-name lookup keyed by the `name` field.
84
84
  */
85
+ /**
86
+ * Stages served by another stage's skill. `testing` shares ct-validator
87
+ * (loomStage `validation`) since ct-ivt-looper was removed (T14311).
88
+ */
89
+ const SHARED_STAGE_SKILL: Readonly<Record<string, string>> = { testing: 'validation' };
90
+
85
91
  const skillByName = new Map(manifest.skills.map((s) => [s.name, s]));
86
92
 
87
93
  // ---------------------------------------------------------------------------
@@ -112,7 +118,8 @@ describe('LOOM stage coverage — dispatch_matrix.by_protocol', () => {
112
118
 
113
119
  describe('LOOM stage coverage — loomStage field on bound skills', () => {
114
120
  for (const stage of CANONICAL_LOOM_STAGES) {
115
- it(`bound skill for "${stage}" declares loomStage === "${stage}"`, () => {
121
+ const expectedLoomStage = SHARED_STAGE_SKILL[stage] ?? stage;
122
+ it(`bound skill for "${stage}" declares loomStage === "${expectedLoomStage}"`, () => {
116
123
  const dashed = stage.replace(/_/g, '-');
117
124
  const skillName =
118
125
  manifest.dispatch_matrix.by_protocol[stage] ??
@@ -121,8 +128,8 @@ describe('LOOM stage coverage — loomStage field on bound skills', () => {
121
128
  expect(skill, `skill "${skillName}" not found in manifest.skills[]`).toBeDefined();
122
129
  expect(
123
130
  skill?.loomStage,
124
- `skill "${skillName}" missing loomStage field; expected "${stage}"`,
125
- ).toBe(stage);
131
+ `skill "${skillName}" missing loomStage field; expected "${expectedLoomStage}"`,
132
+ ).toBe(expectedLoomStage);
126
133
  });
127
134
  }
128
135
  });
@@ -135,8 +142,10 @@ describe('LOOM stage coverage — loomStage values are canonical', () => {
135
142
  const stagesSet = new Set<string>(CANONICAL_LOOM_STAGES);
136
143
  const skillsWithLoomStage = manifest.skills.filter((s) => typeof s.loomStage === 'string');
137
144
 
138
- it('at least 10 skills carry a loomStage field (one per LOOM stage)', () => {
139
- expect(skillsWithLoomStage.length).toBeGreaterThanOrEqual(CANONICAL_LOOM_STAGES.length);
145
+ it('every LOOM stage without a shared skill has its own loomStage skill', () => {
146
+ expect(skillsWithLoomStage.length).toBeGreaterThanOrEqual(
147
+ CANONICAL_LOOM_STAGES.length - Object.keys(SHARED_STAGE_SKILL).length,
148
+ );
140
149
  });
141
150
 
142
151
  for (const skill of skillsWithLoomStage) {
@@ -8,7 +8,7 @@ version: 1
8
8
 
9
9
  protocol subagent-base:
10
10
  description: "RFC 2119 protocol for all CLEO subagents"
11
- version: "1.2.0"
11
+ version: "1.3.0"
12
12
 
13
13
  # --- Required Tokens ---
14
14
 
@@ -37,7 +37,7 @@ protocol subagent-base:
37
37
  # --- Lifecycle Requirements ---
38
38
 
39
39
  constraints [lifecycle]:
40
- LIFE-001: MUST call tasks.start before beginning any work
40
+ LIFE-001: MUST preserve the preallocated worker session and existing task claim; missing, expired or foreign bindings return blocked for coordinator handoff
41
41
  LIFE-002: MUST call tasks.complete after writing output and manifest
42
42
  LIFE-003: MUST NOT call tasks.complete without having written an output file
43
43
 
@@ -61,8 +61,8 @@ protocol subagent-base:
61
61
 
62
62
  phase initialize:
63
63
  step 1: "Read task details via tasks.show ${TASK_ID}"
64
- step 2: "Start task via tasks.start ${TASK_ID}"
65
- LIFE-001: MUST call tasks.start before beginning any work
64
+ step 2: "Verify the preallocated session and claim; export CLEO_SESSION_ID and CLEO_AGENT_ID on every command"
65
+ LIFE-001: MUST NOT start another session or re-claim the task
66
66
 
67
67
  phase execute:
68
68
  step 1: "Follow injected skill protocol for current RCASD-IVTR+C stage"
@@ -90,9 +90,9 @@ protocol subagent-base:
90
90
  - pattern: "Writing pretty-printed JSON to manifest"
91
91
  problem: "Multiple lines break JSONL parsers"
92
92
  solution: "Use cleo manifest append CLI command"
93
- - pattern: "Skipping tasks.start"
94
- problem: "Protocol violation — LIFE-001"
95
- solution: "Always call tasks.start before beginning work"
93
+ - pattern: "Starting or re-claiming a preallocated task"
94
+ problem: "Replaces the worker binding or steals a claim — LIFE-001"
95
+ solution: "Preserve the allocated identity and claim; missing, expired or foreign bindings return blocked for coordinator handoff"
96
96
  - pattern: "Using memory.brain.* prefix"
97
97
  problem: "Removed in ADR-021 — returns E_INVALID_OPERATION"
98
98
  solution: "Use memory.find, memory.observe, memory.link"
@@ -1,8 +1,8 @@
1
1
  # Subagent Protocol Base Reference
2
2
 
3
3
  **Provenance**: @task T3155, @epic T3147
4
- **Version**: 1.1.0
5
- **Updated**: 2026-03-20
4
+ **Version**: 1.3.0
5
+ **Updated**: 2026-11-10
6
6
 
7
7
  This reference defines the RFC 2119 protocol for subagent output and handoff.
8
8
  All subagents operating under an orchestrator MUST follow this protocol.
@@ -97,7 +97,7 @@ See the reference above for:
97
97
 
98
98
  ## Task Lifecycle Integration (LOOM)
99
99
 
100
- All work follows the **LOOM** (Logical Order of Operations Methodology) framework through the RCASD-IVTR+C pipeline stages. See `docs/concepts/CLEO-VISION.md` for the complete LOOM framework, neural hierarchy model, and brain metaphor mapping.
100
+ All work follows the **LOOM** (Logical Order of Operations Methodology) framework through the RCASD-IVTR+C pipeline stages. See `docs/concept/cleo-vision.md` for the complete LOOM framework, neural hierarchy model, and brain metaphor mapping.
101
101
 
102
102
  Reference: @skills/_shared/task-system-integration.md
103
103
 
@@ -105,7 +105,7 @@ Reference: @skills/_shared/task-system-integration.md
105
105
 
106
106
  ```
107
107
  1. Read task: {{TASK_SHOW_CMD}} {{TASK_ID}}
108
- 2. Start task: {{TASK_START_CMD}} {{TASK_ID}}
108
+ 2. Verify the preallocated worker session and existing task claim; preserve both.
109
109
  3. Do work: [skill-specific execution]
110
110
  4. Record output: `cleo docs add {{TASK_ID}} --content - --type <kind> --slug {{TOPIC_SLUG}}`
111
111
  5. Create manifest entry: cleo research add [flags]
@@ -113,6 +113,10 @@ Reference: @skills/_shared/task-system-integration.md
113
113
  7. Return: "[Type] complete. Manifest appended to pipeline_manifest."
114
114
  ```
115
115
 
116
+ Do not start another session or re-claim a preallocated task. Export the allocated
117
+ `CLEO_SESSION_ID` and `CLEO_AGENT_ID` on every command. A missing, expired or foreign
118
+ binding returns blocked to the coordinator for handoff; never replace it yourself.
119
+
116
120
  ---
117
121
 
118
122
  ## Research Linking
@@ -153,7 +157,7 @@ Reference: @skills/_shared/task-system-integration.md
153
157
 
154
158
  Before returning, verify:
155
159
 
156
- - [ ] Task started via `{{TASK_START_CMD}}`
160
+ - [ ] The preallocated worker session and existing claim are preserved
157
161
  - [ ] Output recorded with `cleo docs add` (slug in the return message)
158
162
  - [ ] Manifest entry created via `cleo research add`
159
163
  - [ ] Task completed via `{{TASK_COMPLETE_CMD}}`
@@ -221,5 +225,5 @@ If work cannot proceed:
221
225
  | Returning full content | Bloats orchestrator context | Return only summary message |
222
226
  | Manual JSONL append | No validation, race conditions | Use `cleo research add` |
223
227
  | Missing manifest entry | Orchestrator can't find findings | Always create manifest entry |
224
- | Skipping task start | Protocol violation | Always `cleo start` before work. If `cleo start` fails with `E_SESSION_UNBOUND` (spawned workers and subagents are already bound), run `cleo session start --scope global --name "<name>"` or `cleo session resume <id>` first. |
228
+ | Starting or re-claiming a preallocated task | Replaces the worker binding or steals a claim | Preserve the allocated session and claim. A missing, expired or foreign binding returns blocked to the coordinator for handoff. |
225
229
  | Incomplete checklist | Protocol violation | Verify all items before return |
@@ -6,14 +6,14 @@ adrRefs:
6
6
  - ADR-053
7
7
  - ADR-070
8
8
  metadata:
9
- version: 1.0.2
9
+ version: 1.0.3
10
10
  tier: on-demand
11
11
  install: harness
12
12
  covers:
13
13
  - packages/core/src/validation/protocols/cant/architecture-decision.cant
14
14
  - packages/core/src/validation/protocols/protocols-markdown/architecture-decision.md
15
15
  loomStage: architecture_decision
16
- lastReviewed: 2026-09-28
16
+ lastReviewed: 2026-11-05
17
17
  stability: stable
18
18
  ---
19
19
 
@@ -289,4 +289,4 @@ This skill binds to the **architecture_decision** LOOM lifecycle stage (undersco
289
289
 
290
290
  The stage's canonical name is **`architecture_decision`** (underscored) because the `cleo lifecycle` CLI emits it that way. Historically `packages/skills/skills/manifest.json` `dispatch_matrix.by_protocol` used the dashed form `architecture-decision` — T9672 reconciled that key to the underscored form. The dashed form is retained only as a keyword alias under `dispatch_matrix.by_keyword` so legacy dispatch paths continue to resolve. Frontmatter on this skill (`protocol: architecture_decision`, `loomStage: architecture_decision`) uses the underscored form.
291
291
 
292
- LOOM coverage matrix: [docs/skills/loom-coverage-matrix.md](../../../../docs/skills/loom-coverage-matrix.md).
292
+ LOOM coverage matrix: [docs/reference/loom-coverage-matrix.md](../../../../docs/reference/loom-coverage-matrix.md).
@@ -2,14 +2,14 @@
2
2
  name: ct-cleo
3
3
  description: CLEO task management protocol - session, task, and workflow guidance. Use when managing tasks, sessions, or multi-agent workflows with the CLEO CLI protocol.
4
4
  metadata:
5
- version: 2.24.24
5
+ version: 2.25.9
6
6
  tier: core
7
7
  install: harness
8
8
  covers:
9
9
  - packages/cleo/src/cli/commands/session.ts
10
10
  - packages/cleo/src/cli/commands/focus.ts
11
11
  - packages/cleo/src/cli/commands/sticky.ts
12
- lastReviewed: 2026-10-25
12
+ lastReviewed: 2026-11-10
13
13
  stability: stable
14
14
  ---
15
15
 
@@ -137,8 +137,8 @@ Unattributed mutations warn on stderr; `session status` / `briefing` label a gue
137
137
  session Claude started can be ended from the tab.
138
138
  - Two Claude instances that each start a session stay isolated; a sibling tmux
139
139
  pane never sees another pane's session.
140
- - Agent-tool subagents inherit the parent's `CLAUDE_CODE_SESSION_ID` and act in
141
- the parent's session; `cleo orchestrate spawn` gives workers their own.
140
+ - Agent-tool subagents inherit the parent's `CLAUDE_CODE_SESSION_ID`; each sets
141
+ `CLEO_AGENT_ID` for its own session and claims (else `E_SESSION_AMBIGUOUS`).
142
142
 
143
143
  ## Typed decisions (`cleo decide`, T12491)
144
144
 
@@ -152,7 +152,7 @@ Unattributed mutations warn on stderr; `session status` / `briefing` label a gue
152
152
  | End session (from another terminal) | `cleo session end --session <id>` |
153
153
  | Find work | `cleo next` → `cleo focus <id>` |
154
154
  | Search tasks | `cleo find "query"` |
155
- | Complete task | `cleo verify T### --gate ... --evidence "..."` → `cleo complete T###` |
155
+ | Complete task | PR: `Closes: T### AC1,AC2` in the body, merge on green; the sweep closes it. No PR: `cleo done T### --plan` → `cleo verify …` → `cleo complete T###` |
156
156
  | Run typed gates (the only command that executes them; writes read the cache) | `cleo verify T### --run [--req A,B]` |
157
157
  | Save memory | `cleo memory observe "..." --title "..."` |
158
158
  | Spawn subagent | `cleo orchestrate spawn <taskId> --tier 2` |
@@ -463,4 +463,4 @@ BRAIN ID going forward.
463
463
 
464
464
  ## Evidence must prove task criteria
465
465
 
466
- Merged PRs and passing CI are provenance. Implementation requires changed artifacts related to the task; testing and review require their own actual results (`ci:<pr>` once the PR merged and the project sets `evidence.ciSatisfies`; else `tool:test-affected` when `testing.affectedCommand` is configured, otherwise a targeted `test-run:<json>` or `tool:test`; after merge a scoped run carries only while the change's files are byte-identical, or when it is a `test-run:` recorded on a main HEAD descending from the merge, on the current tree, covering the change (T13673); plan with `cleo done <id> --plan`, which never plans a whole-suite `tool:test` itself (T13647); never run the suite by hand and again via `tool:test`). Run tests, typechecks and builds one at a time through `cleo run --wait --class <test|build|full-build> -- <cmd>` (exit 75: not admitted yet, wait and retry); never wrap `cleo run` or `cleo verify` in another queue or wrapper, and never raise heap or worker counts (`NODE_OPTIONS`, `--maxWorkers`). For tasks with canonical criteria, append explicit links such as `satisfies:T1234#AC1` to each relevant gate's evidence. Fetch the PR merge commit so artifact hashes can be inspected. A changed criterion invalidates its recorded proof. Completing a child preserves an open parent whose own criteria remain unproven; child waivers never transfer to parent criteria.
466
+ A merged PR closes its `Closes:` tasks once `CI` is green on its exact head and a non-author reviewer recorded `met` for each AC (`cleo review verdict`); the merge sweep records `pr:`, `ci:` and `satisfies:`, so no agent step follows the merge (ADR-099; preview with `cleo done --from-merge <pr> --dry-run`). A merge whose own main run failed or was cancelled for a reason that is not the PR's closes once the default branch's head has green CI (D11241): the sweep waits for it, and a failure that is the PR's refuses with the `cleo verify` commands that close the task. No-PR work records its own results: plan with `cleo done <id> --plan` (it never plans a whole-suite `tool:test`, T13647), then the atoms in `cleo briefing inject --section evidence`, each linked with `satisfies:T1234#AC1`. Run tests, typechecks and builds one at a time through `cleo run --wait --class <test|build|full-build> -- <cmd>` (exit 75: not admitted yet, wait and retry); never wrap `cleo run` or `cleo verify` in another queue or wrapper, and never raise heap or worker counts (`NODE_OPTIONS`, `--maxWorkers`). A changed criterion invalidates its recorded proof. Completing a child preserves an open parent whose own criteria remain unproven; child waivers never transfer to parent criteria.
@@ -31,6 +31,12 @@ Use CLI (`cleo`) for all memory operations.
31
31
 
32
32
  - `cleo memory find "{current-topic}" --limit 10`
33
33
  - Avoid re-doing work that's already been completed
34
+ - `cleo memory find "{current-topic}"` also searches the docs store, ranked
35
+ with the BRAIN hits (a full-text index, so it is fast); `--no-docs` leaves
36
+ docs out, `--type doc` searches docs alone, and `--type <brain-type> --docs`
37
+ adds them to one BRAIN table. Only current docs come back: a match on a
38
+ superseded doc returns the doc that replaced it (`replaces` names the old
39
+ slug). Read one with `cleo docs fetch <slug>`.
34
40
 
35
41
  ### After Completing Significant Work
36
42
 
@@ -34,9 +34,13 @@ gates work only where a project turns on the stage gate (below).
34
34
 
35
35
  ## 3. Verify
36
36
 
37
- - `cleo done <id> --plan`, record each gate with `cleo verify <id> --gate <gate>
38
- --evidence "<atoms>"`, then `cleo complete <id>`.
39
- - Each PR head gets an independent review before merge.
37
+ - PR work: the PR body names each task and its ACs (`Closes: T#### AC1,AC2`, or
38
+ `Refs:` for partial delivery). A non-author reviewer records a verdict per AC
39
+ on the PR head (`cleo review verdict`). Merge on green `CI`; the merge sweep
40
+ closes the task. Check the lines first with
41
+ `cleo done --from-merge <pr> --dry-run`.
42
+ - No-PR work: `cleo done <id> --plan`, record each gate with
43
+ `cleo verify <id> --gate <gate> --evidence "<atoms>"`, then `cleo complete <id>`.
40
44
 
41
45
  ## Spawn protocols are labels
42
46
 
@@ -51,5 +55,3 @@ prompt's stage guidance. They are not gates.
51
55
  - Exit 80 naming a planning-stage parent epic: the project turned on
52
56
  `lifecycle.stageGate`; advance the epic before children exist (above), otherwise
53
57
  report it to the lead or owner, who decide on the setting.
54
- - Exit 83 (`E_IVTR_INCOMPLETE`): the task carries `ivtr_state` from the retired
55
- IVTR loop, which no command clears; report it to the lead or owner.
@@ -144,16 +144,14 @@ cleo skills info ct-orchestrator
144
144
 
145
145
  ## Release Workflow
146
146
 
147
- Releases are PR-gated (ADR-065). Commit and merge all code through PRs first; the
148
- release verbs are `plan` → `open` → `reconcile`. The legacy `create`/`ship`
149
- verbs were removed (T9540, T10103).
147
+ A release is one command (ADR-099). Merge all code through PRs first; `cut`
148
+ picks the newest commit on `origin/main` whose `CI` is green, builds the notes
149
+ from the changesets since the last tag, and creates the tag that starts
150
+ `release.yml`. Provenance is reconciled automatically by the merge sweep.
150
151
 
151
152
  ```bash
152
- cleo release plan v2026.MM.N --epic T#### # or --tasks T1,T2
153
- cleo release open v2026.MM.N # dispatches release-prepare
154
- cleo release pr-status v2026.MM.N # poll the release PR + CI
155
- git tag -a v2026.MM.N -m "Release v2026.MM.N" && git push origin v2026.MM.N
156
- cleo release reconcile v2026.MM.N # backfill provenance
153
+ cleo release cut v2026.MM.N --dry-run # sha, CI run, notes, packages, task report
154
+ cleo release cut v2026.MM.N # tag and publish
157
155
  ```
158
156
 
159
157
  ## Project Context
@@ -1,6 +1,6 @@
1
1
  ---
2
2
  name: ct-codebase-mapper
3
- version: 2.3.5
3
+ version: 2.4.0
4
4
  description: Orient in an unfamiliar or large codebase with CLEO's code-intelligence graph (cleo nexus) and project map (cleo map). Use before planning or editing unfamiliar code, for brownfield onboarding, to find what a change would break, or to map a project's structure, communities and execution flows. Triggers on "map the codebase", "understand this project", "what calls X", "what would break", "brownfield analysis", "project structure".
5
5
  protocol: null
6
6
  dependencies: []
@@ -17,13 +17,13 @@ triggers:
17
17
  - project structure
18
18
  - blast radius
19
19
  metadata:
20
- version: 2.3.5
20
+ version: 2.4.0
21
21
  tier: on-demand
22
22
  install: harness
23
23
  covers:
24
24
  - packages/cleo/src/cli/commands/nexus.ts
25
25
  - packages/cleo/src/cli/commands/map.ts
26
- lastReviewed: 2026-10-24
26
+ lastReviewed: 2026-11-09
27
27
  stability: stable
28
28
  ---
29
29
 
@@ -118,6 +118,29 @@ caller, so an empty footprint alone never establishes that a change is safe.
118
118
  When a symbol name is ambiguous, pass the qualified candidate the command
119
119
  returns and confirm against the source.
120
120
 
121
+ `impact` judges `coverage` by the files its answer rests on: the target's file
122
+ and every traversed caller's file. A file that failed extraction elsewhere
123
+ makes the answer `partial` (one reason counts those files; callers in them may
124
+ be missing) instead of hiding the callers it found; a failed target or caller
125
+ file makes it `failed`. Any status short of `current` gives `UNKNOWN`. The
126
+ reply carries a bounded coverage sample (`reasonCount`, `detailsCommand`);
127
+ `--coverage-detail` returns every reason and limitation, including unrelated
128
+ files.
129
+
130
+ ## 5. Which tests a change reaches
131
+
132
+ ```bash
133
+ cleo nexus affected --base origin/main # test files, vitest projects and changed symbols
134
+ ```
135
+
136
+ It walks the static import graph of the checkout (not the NEXUS index, so it
137
+ is never stale): from each changed module through its importers, re-exports
138
+ followed by name, to the test files that reach it. `mode: full` with a
139
+ `reason` means the graph cannot speak for the change (a migration, fixture or
140
+ template, a root config, a module no test reaches); select by package or run
141
+ the full suite then. PR CI's unit shards and `tool:test-affected` use the same
142
+ selection.
143
+
121
144
  On a stale index, `impact` and `context` refresh it inline when the size-scaled
122
145
  cost estimate fits the budget: one refresh at a time per project, a larger one
123
146
  only when `cleo run` admits it, and none for 30 minutes after one fails or is
@@ -163,6 +186,7 @@ concluding "none".
163
186
  | Planning an epic in unfamiliar code | `cleo map --focus architecture`, `cleo nexus clusters` |
164
187
  | About to change a function or type | `cleo nexus impact <symbol>` |
165
188
  | "Who calls this / what does this call?" | `cleo nexus context <symbol>` |
189
+ | "Which tests does my change reach?" | `cleo nexus affected --base origin/main` |
166
190
  | Tech-debt review | `cleo map --focus concerns` |
167
191
  | "Which machines hold this project, and is it pushed?" | `cleo nexus projects status` |
168
192
 
@@ -6,14 +6,14 @@ adrRefs:
6
6
  - ADR-015
7
7
  - ADR-023
8
8
  metadata:
9
- version: 1.0.2
9
+ version: 1.0.3
10
10
  tier: on-demand
11
11
  install: harness
12
12
  covers:
13
13
  - packages/core/src/validation/protocols/cant/consensus.cant
14
14
  - packages/core/src/validation/protocols/protocols-markdown/consensus.md
15
15
  loomStage: consensus
16
- lastReviewed: 2026-09-28
16
+ lastReviewed: 2026-11-05
17
17
  stability: stable
18
18
  ---
19
19
 
@@ -177,4 +177,4 @@ This skill binds to the **consensus** LOOM lifecycle stage. Governing ADRs:
177
177
  - [ADR-015 — multi-contributor architecture](../../../../.cleo/adrs/ADR-015-multi-contributor-architecture.md) — defines the consensus framework that this skill implements.
178
178
  - [ADR-023 — protocol validation dispatch](../../../../.cleo/adrs/ADR-023-protocol-validation-dispatch.md) — defines how consensus output is validated before downstream stages consume it.
179
179
 
180
- LOOM coverage matrix: [docs/skills/loom-coverage-matrix.md](../../../../docs/skills/loom-coverage-matrix.md).
180
+ LOOM coverage matrix: [docs/reference/loom-coverage-matrix.md](../../../../docs/reference/loom-coverage-matrix.md).
@@ -6,7 +6,7 @@ description: >-
6
6
  "consensus workflow", "multi-agent decision", "create contribution",
7
7
  "contribution start", "contribution submit", "detect conflicts",
8
8
  "weighted consensus", "decision tracking", "conflict resolution".
9
- version: 1.0.2
9
+ version: 1.0.3
10
10
  protocol: contribution
11
11
  adrRefs:
12
12
  - ADR-015
@@ -21,14 +21,14 @@ compatibility:
21
21
  - gemini-cli
22
22
  license: MIT
23
23
  metadata:
24
- version: 1.0.2
24
+ version: 1.0.3
25
25
  tier: on-demand
26
26
  install: harness
27
27
  covers:
28
28
  - packages/core/src/validation/protocols/cant/contribution.cant
29
29
  - packages/core/src/validation/protocols/protocols-markdown/contribution.md
30
30
  loomStage: contribution
31
- lastReviewed: 2026-09-28
31
+ lastReviewed: 2026-11-05
32
32
  stability: stable
33
33
  ---
34
34
 
@@ -604,4 +604,4 @@ This skill binds to the **contribution** LOOM lifecycle stage (the final stage o
604
604
  - [ADR-015 — multi-contributor architecture](../../../../.cleo/adrs/ADR-015-multi-contributor-architecture.md) — defines the multi-contributor consensus mechanics that this stage formalizes for an Epic's downstream return path.
605
605
  - [ADR-053 — playbook runtime](../../../../.cleo/adrs/ADR-053-playbook-runtime.md) — defines the lifecycle state machine; contribution is its terminal node.
606
606
 
607
- LOOM coverage matrix: [docs/skills/loom-coverage-matrix.md](../../../../docs/skills/loom-coverage-matrix.md).
607
+ LOOM coverage matrix: [docs/reference/loom-coverage-matrix.md](../../../../docs/reference/loom-coverage-matrix.md).