@event4u/agent-config 6.0.0 → 6.1.0

package/dist/agent-src/commands/memory/add.md

@@ -1,11 +1,12 @@
 ---
 model_tier: medium
 name: memory-add
-pack: meta
+pack: memory
 tier: 2
+visibility: internal
 cluster: memory
 sub: add
-description: Interactively add a validated entry to an engineering-memory file (domain-invariants, architecture-decisions, incident-learnings, product-rules)
+description: Interactively add a validated entry to an engineering-memory file (domain-invariants, incident-learnings, product-rules, ownership, historical-patterns)
 skills: [file-editor]
 suggestion:
   eligible: true
@@ -14,7 +15,7 @@ suggestion:
 workspaces:
   - agent-config-maintainer
 packs:
-  - meta
+  - memory
 ---
 
 # /memory add
@@ -41,9 +42,10 @@ the first write and for every subsequent addition.
 > Which memory type is this entry for?
 >
 > 1. domain-invariants  — architectural boundaries, module rules
-> 2. architecture-decisions — ADR-style decisions with alternatives
-> 3. incident-learnings — post-mortem patterns with guardrails
-> 4. product-rules — intentional business constraints
+> 2. incident-learnings — post-mortem patterns with guardrails
+> 3. product-rules — intentional business constraints
+>
+> (Architecture decisions belong in ADRs — `docs/decisions/` — not memory.)
 ```
 
 If the user cannot pick one confidently, stop and ask what the entry

package/dist/agent-src/commands/memory/learn-low-impact.md

@@ -1,8 +1,9 @@
 ---
 model_tier: medium
 name: memory-learn-low-impact
-pack: meta
+pack: memory
 tier: 2
+visibility: internal
 cluster: memory
 sub: learn-low-impact
 skills: [ai-council, upstream-contribute]
@@ -14,7 +15,7 @@ suggestion:
 workspaces:
   - agent-config-maintainer
 packs:
-  - meta
+  - memory
 ---
 
 # /memory learn-low-impact

package/dist/agent-src/commands/memory/load.md

@@ -1,8 +1,9 @@
 ---
 model_tier: medium
 name: memory-load
-pack: meta
+pack: memory
 tier: 2
+visibility: internal
 cluster: memory
 sub: load
 description: Load ALL curated entries of a given memory type into the current context — opt-in full load for deep analysis, never auto-triggered
@@ -13,7 +14,7 @@ suggestion:
 workspaces:
   - agent-config-maintainer
 packs:
-  - meta
+  - memory
 ---
 
 # /memory load
@@ -40,11 +41,10 @@ in the entry count and the signal-to-noise degrades fast on large types.
 > Which memory type should I load in full?
 >
 > 1. domain-invariants
-> 2. architecture-decisions
-> 3. incident-learnings
-> 4. product-rules
-> 5. historical-patterns
-> 6. ownership
+> 2. incident-learnings
+> 3. product-rules
+> 4. historical-patterns
+> 5. ownership
 ```
 
 ### 1b. Surface the Tier-0 critical slice

package/dist/agent-src/commands/memory/mine-session.md

@@ -1,11 +1,12 @@
 ---
 model_tier: medium
 name: memory-mine-session
-pack: meta
+pack: memory
 tier: 2
+visibility: internal
 cluster: memory
 sub: mine-session
-description: Mine the active session transcript for memory signals (corrections, preferences, decisions, recurring patterns) — preview-by-default, opt-in transcript access, host-agnostic via TranscriptAdapter.
+description: Mine a session (cross-host chat-history log) for memory signals and/or rule/skill proposal seeds via --mode=[signals|proposals|both]. Preview-default, opt-in. Folds in /chat-history learn.
 skills: [memory-consolidation, file-editor]
 suggestion:
   eligible: false
@@ -13,15 +14,28 @@ suggestion:
 workspaces:
   - agent-config-maintainer
 packs:
-  - meta
+  - memory
 ---
 
 # /memory mine-session
 
-Runs the **GATHER SIGNAL** phase of the [`memory-consolidation`](../../skills/memory-consolidation/SKILL.md)
-loop against the current host's transcripts and surfaces normalised
-project-scoped facts as a preview. No file is written without
-`--commit-intake`. No transcript is read without
+The **single session-mining command**. Runs the **GATHER SIGNAL** phase of
+the [`memory-consolidation`](../../skills/memory-consolidation/SKILL.md) loop
+against the **cross-host chat-history JSONL log**
+(`agents/runtime/.agent-chat-history`, written by platform hooks on every
+host) — falling back to the per-host Claude-Code transcript when the log is
+absent — and surfaces normalised project-scoped facts as a preview.
+
+`--mode` selects the output:
+
+- **`signals`** (default) — normalised facts → intake preview / `--commit-intake`.
+- **`proposals`** — frames the facts as candidate rule/skill learnings, then
+  run [`learning-to-rule-or-skill`](../../skills/learning-to-rule-or-skill/SKILL.md)
+  on each seed to draft proposals. **This folds in the former
+  `/chat-history learn`.**
+- **`both`** — render signals + proposal seeds together.
+
+No file is written without `--commit-intake`. No source is read without
 `--confirm-transcript-access`.
 
 ## When to use
@@ -41,11 +55,13 @@ and confirmed.
 
 | Flag | Default | Purpose |
 |---|---|---|
+| `--mode <signals\|proposals\|both>` | `signals` | Output selector. `signals` → intake facts; `proposals` → proposal seeds for `learning-to-rule-or-skill`; `both` → both. |
+| `--source <auto\|chat-history\|claude-code>` | `auto` | `auto` prefers the cross-host chat-history log, then the Claude-Code transcript. |
 | `--since <ISO-date>` | 14 days ago | Only mine turns at or after this date. Re-anchors relative phrases like "yesterday" with `YYYY-MM-DD`. |
 | `--confirm-transcript-access` | off | Opt-in gate. Without it, the miner reads zero turns and prints the opt-in hint. Per-invocation, not persistent. |
-| `--preview` | on | Render normalised facts to stdout. No file write. Default behaviour. |
-| `--commit-intake` | off | Mutually exclusive with `--preview`. Append normalised facts to `agents/memory/intake/<primary-tag>.jsonl`. |
-| `--host <claude-code\|cursor\|augment>` | auto-detect | Force the `TranscriptAdapter`. Phase 1 ships `claude-code` only; others print `not-supported-on-this-host` and exit. |
+| `--preview` | on | Render to stdout. No file write. Default behaviour. |
+| `--commit-intake` | off | `signals` only. Append normalised facts to `agents/memory/intake/<primary-tag>.jsonl`. |
+| `--host <name>` | auto-detect | Labels the source host. The chat-history log is cross-host, so any host mines via it; `claude-code` additionally resolves the local transcript fallback. |
 
 ## Steps
 
@@ -91,6 +107,17 @@ If the normalised count exceeds 5, **stop after the 5th** and print:
 
 This is the strict-gate exit per the skill's exit criteria.
 
+**Failure-mode review lens** (adapted from MemSkill, Apache-2.0). Before
+emitting a fact, ask which failure it prevents — and drop it if it prevents
+none:
+
+- **Storage failure** — a durable fact the next agent will need was never
+  written. Capture it.
+- **Retrieval failure** — the fact exists but its `key` / anchor path won't
+  match how a future skill queries. Fix the anchor, don't duplicate.
+- **Quality failure** — the fact is too vague or narrative to act on. Sharpen
+  it to one durable PATTERN / INVARIANT / GOTCHA, or drop it.
+
 ### 4. Render preview
 
 Print one Markdown block to stdout:
@@ -153,5 +180,5 @@ Confirm:
   `TranscriptAdapter` matches the current host.
 - [`memory:promote`](promote.md) — lifts validated intake lines into
   curated YAML.
-- [`agent-memory-contract`](../../../docs/contracts/agent-memory-contract.md) —
-  intake JSONL schema, including the `tags` field.
+- [`memory-access`](../../../docs/guidelines/agent-infra/memory-access.md) —
+  file-backed retrieval contract for curated + intake entries.

package/dist/agent-src/commands/memory/promote.md

@@ -1,8 +1,9 @@
 ---
 model_tier: medium
 name: memory-promote
-pack: meta
+pack: memory
 tier: 2
+visibility: internal
 cluster: memory
 sub: promote
 description: Promote an intake signal (or provisional proposal) into a curated memory entry — opens a PR and runs the admission gate.
@@ -13,7 +14,7 @@ suggestion:
 workspaces:
   - agent-config-maintainer
 packs:
-  - meta
+  - memory
 ---
 
 # /memory promote

package/dist/agent-src/commands/memory/propose.md

@@ -1,8 +1,9 @@
 ---
 model_tier: medium
 name: memory-propose
-pack: meta
+pack: memory
 tier: 2
+visibility: internal
 cluster: memory
 sub: propose
 description: Append a provisional memory signal to the intake stream — the universal fallback for any producer (human or agent) to record a finding without committing to a curated entry.
@@ -13,7 +14,7 @@ suggestion:
 workspaces:
   - agent-config-maintainer
 packs:
-  - meta
+  - memory
 ---
 
 # /memory propose
@@ -35,8 +36,9 @@ not yet justify a human reviewer's attention.
 
 Do NOT use when:
 
-- The entry is a **decision** — go to [`/memory-add`](memory-add.md)
-  with `architecture-decisions` or `product-rules`.
+- The entry is a **product decision** — go to [`/memory-add`](memory-add.md)
+  with `product-rules`. (Architecture decisions belong in ADRs —
+  `docs/decisions/` — not memory.)
 - The finding contradicts an existing curated entry — open a
   supersede-chain discussion instead.
 
@@ -51,8 +53,7 @@ Do NOT use when:
 > 2. incident-learnings  — a post-mortem guardrail
 > 3. ownership           — a path → team mapping
 > 4. domain-invariants   — a module boundary rule
-> 5. architecture-decisions — a decision to revisit
-> 6. product-rules       — an intentional constraint
+> 5. product-rules       — an intentional constraint
 ```
 
 ### 2. Gather minimum fields

package/dist/agent-src/commands/memory.md

@@ -2,11 +2,12 @@
 model_tier: medium
 name: memory
 disable-model-invocation: true
-pack: meta
+pack: memory
 intent: "Manage the agent memory layer — add, load, promote, propose"
 routes_to: [memory-add, memory-load, memory-promote, memory-propose]
 replaces: []
 tier: 1
+visibility: advanced
 description: Memory orchestrator — routes to add, load, mine-session, promote, propose
 cluster: memory
 type: orchestrator
@@ -17,7 +18,7 @@ suggestion:
 workspaces:
   - agent-config-maintainer
 packs:
-  - meta
+  - memory
 ---
 
 # /memory

package/dist/agent-src/commands/mode.md

@@ -6,6 +6,7 @@ intent: "Switch the active role-mode (developer/reviewer/tester/po/incident/plan
 routes_to: [role-mode-adherence]
 replaces: []
 tier: 1
+visibility: advanced
 description: Set the active role mode — prints the contract, lists default skills, and refuses work outside the contract (see role-contracts)
 suggestion:
   eligible: false

package/dist/agent-src/commands/module/create.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: module-create
 pack: engineering-base
 tier: 2
+visibility: internal
 cluster: module
 sub: create
 skills: [laravel]

package/dist/agent-src/commands/module/explore.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: module-explore
 pack: engineering-base
 tier: 2
+visibility: internal
 cluster: module
 sub: explore
 skills: [laravel, symfony-workflow, php-coder]

package/dist/agent-src/commands/module.md

@@ -4,6 +4,7 @@ name: module
 disable-model-invocation: true
 pack: engineering-base
 tier: 2
+visibility: internal
 description: Module orchestrator — routes to create, explore
 cluster: module
 type: orchestrator

package/dist/agent-src/commands/optimize/agents-dir.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: optimize-agents-dir
 pack: meta
 tier: 2
+visibility: internal
 cluster: optimize
 sub: agents-dir
 description: Manage the agents/ directory — scaffold, folder-audit, fix. Single command with three modes (--scaffold / --audit / --fix); default = interactive wizard.

package/dist/agent-src/commands/optimize/augmentignore.md

@@ -4,6 +4,7 @@ skills: [agent-docs-writing]
 name: optimize-augmentignore
 pack: meta
 tier: 2
+visibility: internal
 cluster: optimize
 sub: augmentignore
 description: Creates or updates .augmentignore based on the project's actual tech stack, large files, generated artifacts, and irrelevant agent skills/rules.

package/dist/agent-src/commands/optimize/rtk.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: optimize-rtk
 pack: meta
 tier: 2
+visibility: internal
 cluster: optimize
 sub: rtk
 skills: [agent-docs-writing]

package/dist/agent-src/commands/optimize/skills.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: optimize-skills
 pack: meta
 tier: 2
+visibility: internal
 cluster: optimize
 sub: skills
 description: Audits skills — measures baseline, finds duplicates/merge candidates, runs linter. Suggest only, never auto-apply.

package/dist/agent-src/commands/optimize-prompt.md

@@ -3,6 +3,7 @@ model_tier: inherit
 name: optimize-prompt
 pack: meta
 tier: 2
+visibility: internal
 cluster: optimize
 description: "Optimize a raw prompt for ChatGPT, Claude, Gemini, or another AI via the 4-D methodology — BASIC vs DETAIL auto-detect, one clarifying question per turn, returns the polished prompt."
 skills: [prompt-optimizer]

package/dist/agent-src/commands/optimize.md

@@ -7,6 +7,7 @@ intent: "Optimization dispatcher — skills, rtk, augmentignore, agents-dir"
 routes_to: [optimize-skills, optimize-rtk, optimize-augmentignore, optimize-agents-dir]
 replaces: []
 tier: 1
+visibility: advanced
 description: Optimize orchestrator — routes to skills, agents-dir, augmentignore, rtk-filters
 cluster: optimize
 type: orchestrator

package/dist/agent-src/commands/orchestrate.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: orchestrate
 pack: meta
 tier: 2
+visibility: internal
 cluster: orchestrate
 skills: [subagent-orchestration]
 description: Run a YAML pipeline defined under `.agent-config/orchestrations/` — chains personas / skills / commands / sub-agents per the orchestration-dsl-v1 contract

package/dist/agent-src/commands/override/create.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: override-create
 pack: meta
 tier: 2
+visibility: internal
 cluster: override
 sub: create
 description: Creates a project-level override for a shared skill, rule, or command.

package/dist/agent-src/commands/override/manage.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: override-manage
 pack: meta
 tier: 2
+visibility: internal
 cluster: override
 sub: manage
 description: Reviews, updates, and refactors existing project-level overrides.

package/dist/agent-src/commands/override.md

@@ -4,6 +4,7 @@ name: override
 disable-model-invocation: true
 pack: meta
 tier: 2
+visibility: internal
 description: Override orchestrator — routes to create, manage
 cluster: override
 type: orchestrator

package/dist/agent-src/commands/package-reset.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: package-reset
 pack: meta
 tier: 2
+visibility: internal
 skills: [composer, npm]
 description: /package-reset
 suggestion:

package/dist/agent-src/commands/package-test.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: package-test
 pack: engineering-base
 tier: 2
+visibility: internal
 skills: [composer, npm, python-packages, go-modules, cargo-packages]
 description: /package-test
 suggestion:

package/dist/agent-src/commands/post-as/ghostwriter.md

@@ -3,6 +3,7 @@ model_tier: inherit
 name: post-as-ghostwriter
 pack: gtm-marketing
 tier: 2
+visibility: internal
 cluster: post-as
 sub: ghostwriter
 skills: [post-as, ghostwriter]

package/dist/agent-src/commands/post-as/me.md

@@ -3,6 +3,7 @@ model_tier: inherit
 name: post-as-me
 pack: gtm-marketing
 tier: 2
+visibility: internal
 cluster: post-as
 sub: me
 skills: [post-as]

package/dist/agent-src/commands/post-as.md

@@ -3,6 +3,7 @@ model_tier: inherit
 name: post-as
 pack: gtm-marketing
 tier: 2
+visibility: internal
 description: Consumer-facing write entry points — :me drafts in the maintainer's own voice from .agent-user.md (no disclosure); :ghostwriter is a thin alias for /ghostwriter:write (mandatory disclosure footer).
 cluster: post-as
 type: orchestrator

package/dist/agent-src/commands/pr/create/description-only.md

@@ -4,6 +4,7 @@ name: git-pr-create-description-only
 pack: git
 replaces: [create-pr-description-only, create-pr:description-only]
 tier: 2
+visibility: internal
 cluster: git-pr-create
 sub: description-only
 skills: [git-workflow]

package/dist/agent-src/commands/pr/create.md

@@ -7,6 +7,7 @@ intent: "Open a pull request with a generated description and stripped attributi
 routes_to: [git-workflow]
 replaces: [create-pr]
 tier: 1
+visibility: advanced
 cluster: git-pr-create
 skills: [git-workflow]
 description: Create a GitHub PR with structured description from Jira ticket and code changes
@@ -80,6 +81,30 @@ Never improvise the base or silently proceed when the branch is behind
 and overlap exists. The 10-second fetch beats hours of rebase
 reconciliation after the parent PR lands.
 
+### 1c. PR-gate — archive completed roadmaps (MANDATORY)
+
+A roadmap that reached 100% (`count_open == 0 && count_deferred == 0`) must
+land **already archived** in this PR — never merged-but-unarchived into the
+trunk (that rot is exactly what the PR-gate replaces; see
+[`roadmap-progress-sync`](../../../rules/roadmap-progress-sync.md) § PR-gate).
+Run the deterministic sweep:
+
+```bash
+python3 .augment/scripts/archive_completed_roadmaps.py   # --changed-only (default)
+```
+
+It `git mv`s each completed roadmap (that this branch touched, per
+`git log origin/main..HEAD`) into `agents/roadmaps/archive/`, rewrites inbound
+`agents/roadmaps/<x>.md` references to the archive path, regenerates the
+dashboard, and stages all of it. No agent-set annotation is involved —
+completion is read from the checkbox counts.
+
+- **It staged changes → commit them onto this branch** (`chore(roadmaps):
+  archive completed roadmaps`) and include them in the push, so the PR carries
+  the archival. Do **not** create the PR with completed-but-unarchived roadmaps
+  in the working tree.
+- **It reported nothing → proceed.** No completed roadmap in this branch.
+
 ### 2. Generate PR content
 
 Run `/create-pr:description-only` Steps 1–4 to generate the PR title and body.

package/dist/agent-src/commands/prediction-pool.md

@@ -3,6 +3,7 @@ model_tier: inherit
 name: prediction-pool
 pack: fun
 tier: 2
+visibility: internal
 description: "Fill a prediction pool (kicktipp, football/basketball WM): optimize expected points under the rules, enter tips via Playwright. Triggers 'Tippspiel', 'kicktipp', 'predict the pool'."
 skills: [prediction-pool-optimizer]
 suggestion:

package/dist/agent-src/commands/prepare-for-review.md

@@ -3,6 +3,7 @@ model_tier: high
 name: prepare-for-review
 pack: engineering-base
 tier: 2
+visibility: internal
 skills: [git-workflow, quality-tools]
 description: Prepare a PR branch for local review — updates main and merges the full branch chain so the branch is up to date
 suggestion:

package/dist/agent-src/commands/profile/activate.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: profile-activate
 pack: meta
 tier: 2
+visibility: internal
 cluster: profile
 sub: activate
 description: Activate a session profile — surface only the named profile/pack closure plus core artefacts, no persistence

package/dist/agent-src/commands/profile/deactivate.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: profile-deactivate
 pack: meta
 tier: 2
+visibility: internal
 cluster: profile
 sub: deactivate
 description: Deactivate the session profile — clear the overlay (or drop named packs) so the full surface returns

package/dist/agent-src/commands/profile/show.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: profile-show
 pack: meta
 tier: 2
+visibility: internal
 cluster: profile
 sub: show
 description: Show the active session profile — active packs and surfaced/hidden command+skill counts (observability surface)

package/dist/agent-src/commands/profile.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: profile
 pack: meta
 tier: 2
+visibility: internal
 description: Session-profile orchestrator — activate / deactivate / show the active packs for this session (recommendation-bias surface filter, no persistence)
 cluster: profile
 type: orchestrator

package/dist/agent-src/commands/project-analyze.md

@@ -3,6 +3,7 @@ model_tier: high
 name: project-analyze
 pack: engineering-base
 tier: 2
+visibility: internal
 skills: [project-analyzer]
 description: Full project analysis — detect stack, inventory modules, audit docs, create missing contexts
 suggestion:

package/dist/agent-src/commands/project-health.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: project-health
 pack: engineering-base
 tier: 2
+visibility: internal
 skills: [quality-tools]
 description: Quick project health check — show status of docs, modules, contexts, and roadmaps without creating anything
 suggestion:

package/dist/agent-src/commands/quality-fix.md

@@ -3,6 +3,7 @@ model_tier: medium
 name: quality-fix
 pack: engineering-base
 tier: 2
+visibility: internal
 skills: [quality-tools]
 description: Run quality pipeline (PHP and/or JS/TS) and fix all errors — auto-detects language from changed files
 suggestion:

package/dist/agent-src/commands/refine-ticket.md

@@ -6,6 +6,7 @@ intent: "Refine a vague ticket into a clear, implementable spec"
 routes_to: [refine-ticket]
 replaces: []
 tier: 1
+visibility: advanced
 skills: [refine-ticket]
 description: Refine a Jira/Linear ticket before planning — rewritten ticket + Top-5 risks + persona voices, orchestrates validate-feature-fit and threat-modeling, ends with a close-prompt
 suggestion:

package/dist/agent-src/commands/research/deep.md

@@ -3,6 +3,7 @@ model_tier: high
 name: research-deep
 pack: product-discovery
 tier: 2
+visibility: internal
 cluster: research
 sub: deep
 description: "Read `outline.yaml`, research each item in batches, write per-item JSON validated against the project-local research-schema. No Python runtime, no `~/.claude/` paths."

package/dist/agent-src/commands/research/report.md

@@ -3,6 +3,7 @@ model_tier: high
 name: research-report
 pack: product-discovery
 tier: 2
+visibility: internal
 cluster: research
 sub: report
 description: "Summarise per-item JSON results from `/research:deep` into `report.md`. Agent renders directly + emits an optional `jq` template for deterministic regeneration. No Python runtime."

package/dist/agent-src/commands/research.md

@@ -4,6 +4,7 @@ name: research
 disable-model-invocation: true
 pack: product-discovery
 tier: 2
+visibility: internal
 cluster: research
 description: "Preliminary research scaffolder — pick objects, define fields, emit `outline.yaml` + `fields.yaml` for downstream deep research. Use for surveys, benchmarks, tech selection, competitive scans."
 skills: [project-analyzer, deep-reading-analyst]

package/dist/agent-src/commands/review-changes.md

@@ -6,6 +6,7 @@ intent: "Multi-judge review of the current diff — bugs, security, tests, quali
 routes_to: [code-review, judge-bug-hunter, judge-security-auditor, judge-test-coverage, judge-code-quality, architecture-review-lens]
 replaces: []
 tier: 1
+visibility: advanced
 skills: [code-review, subagent-orchestration, judge-bug-hunter, judge-security-auditor, judge-test-coverage, judge-code-quality, architecture-review-lens, git-workflow]
 description: Self-review local changes before creating a PR — dispatches to five specialized judges (bug, security, tests, quality, architecture) and consolidates verdicts
 suggestion:

package/dist/agent-src/commands/review-routing.md

@@ -3,6 +3,7 @@ model_tier: high
 name: review-routing
 pack: meta
 tier: 2
+visibility: internal
 skills: [review-routing, reviewer-awareness]
 description: Compute reviewer roles and matched historical bug patterns for the current diff, using project-local ownership-map.yml and historical-bug-patterns.yml
 suggestion:

package/dist/agent-src/commands/roadmap/ai-council.md

@@ -3,6 +3,7 @@ model_tier: inherit
 name: roadmap-ai-council
 pack: product-basic
 tier: 2
+visibility: internal
 cluster: roadmap
 sub: ai-council
 skills: [ai-council, agent-docs-writing, roadmap-management]

package/dist/agent-src/commands/roadmap/create.md

@@ -3,6 +3,7 @@ model_tier: inherit
 name: roadmap-create
 pack: product-basic
 tier: 2
+visibility: internal
 cluster: roadmap
 sub: create
 skills: [agent-docs-writing]