@event4u/agent-config 6.1.0 → 7.0.0

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

@@ -0,0 +1,124 @@
+---
+model_tier: medium
+name: analyze
+disable-model-invocation: true
+pack: analysis-workbench
+intent: "Analysis dispatcher — classify input by keywords, propose a weighted framework path, let the user pick"
+routes_to: [analyze-postmortem, analyze-premortem, analyze-decision, analyze-near-miss, analyze-incident]
+replaces: []
+tier: 1
+visibility: advanced
+description: Analysis orchestrator — confidence-weighted suggester that routes to postmortem, premortem, decision-review, near-miss, or incident frameworks.
+cluster: analyze
+type: orchestrator
+suggestion:
+  eligible: true
+  trigger_description: "analyze this, run a post-mortem, pre-mortem, decision review, near-miss, or incident analysis"
+  trigger_context: "user wants to apply a structured analysis framework to a past event, future risk, or past decision"
+workspaces:
+  - engineering
+packs:
+  - analysis-workbench
+---
+
+# /analyze
+
+Top-level orchestrator for the `/analyze` family. A
+**confidence-weighted suggester** — classifies the user's input by
+keywords, proposes a numbered-options menu ranked by fit, and routes
+to the chosen sub-command. Never auto-selects silently.
+
+## Sub-commands
+
+| Sub-command | Routes to | Purpose |
+|---|---|---|
+| `/analyze:postmortem` | `analyze/postmortem/command.md` | Blame-free post-mortem after a resolved incident |
+| `/analyze:premortem` | `analyze/premortem/command.md` | Forward-looking imagined-failure before committing to a plan |
+| `/analyze:decision` | `analyze/decision/command.md` | Audit a past architectural decision (did it hold?) |
+| `/analyze:near-miss` | `analyze/near-miss/command.md` | Post-mortem for a near-miss (same flow, lighter severity framing) |
+| `/analyze:incident` | `analyze/incident/command.md` | Full incident flow: commander → RCA → post-mortem → memory candidate |
+
+## Dispatch
+
+### 1. Parse the argument
+
+`/analyze[:<sub>] [args]` or `/analyze <sub> [args]`.
+
+- **Explicit sub-command present** → load that sub-command's body and
+  follow its `## Instructions` verbatim with the remaining args.
+- **No sub-command** → proceed to Step 2.
+
+### 2. Confidence-weighted classification
+
+Read the user's free-text input. Score each framework by keyword
+match:
+
+| Keyword signals | Suggests |
+|---|---|
+| "post-mortem", "outage", "incident resolved", "was ist schiefgelaufen" | `postmortem` |
+| "near-miss", "almost failed", "beinahe-Vorfall", "close call" | `near-miss` |
+| "pre-mortem", "what could go wrong", "imagine failure", "vor dem Start" | `premortem` |
+| "decision review", "ADR", "did this hold up", "rückblick Architektur" | `decision` |
+| "production is down", "active incident", "Vorfall", "Prod ist down" | `incident` |
+
+Assign a confidence level (high / medium / low) per framework based on
+signal count. A framework with zero signals gets `low`.
+
+### 3. Surface the weighted menu — NEVER auto-select
+
+Present numbered options in descending confidence order. At least two
+options must appear. Always include the `describe in your own words`
+escape:
+
+> Which analysis framework fits?
+>
+> 1. [high] Post-mortem (`/analyze:postmortem`) — blame-free write-up after a resolved incident
+> 2. [medium] Near-miss (`/analyze:near-miss`) — same flow, lighter severity framing
+> 3. [low] Pre-mortem (`/analyze:premortem`) — imagined-failure analysis before committing
+> 4. [low] Decision review (`/analyze:decision`) — did a past architectural decision hold up?
+> 5. [low] Incident (`/analyze:incident`) — full live-incident coordination + RCA + post-mortem
+> 6. Describe in your own words — I'll re-classify
+
+**Empfehlung:** 1
+
+The user picks a number. Load the routed sub-command and follow its
+`## Instructions` verbatim with the remaining args.
+
+### 4. If the user picks "Describe in your own words"
+
+Ask one question (per `ask-when-uncertain`):
+
+> What happened or what are you trying to analyse?
+
+Re-run Step 2 with the answer. Present the menu again. Do not loop
+more than twice — on the third attempt, show all five options unranked
+and let the user pick.
+
+### 5. Unknown sub-command
+
+If the argument names a sub-command not in the table, show the table
+and ask which one was intended. One question, no guessing.
+
+## Rules
+
+- **Never auto-select a framework** — always show the weighted menu
+  and wait for the user's pick (council-locked design, Phase 4).
+- **Explicit routing only in v1** — do not infer the framework from
+  open files, branch names, or context not supplied by the user in
+  this turn.
+- **Do NOT commit, push, or open a PR** unless a sub-command
+  explicitly authorizes it.
+- **Do NOT chain sub-commands.** One `/analyze <sub>` per turn.
+- **Memory loop applies** — every sub-command that produces an
+  analysis output MUST draft a memory candidate via
+  `/memory propose` per
+  [`docs/contracts/analysis-memory-loop.md`](../docs/contracts/analysis-memory-loop.md).
+
+## See also
+
+- [`docs/contracts/analysis-memory-loop.md`](../docs/contracts/analysis-memory-loop.md) — produce → propose → promote → retrieve contract
+- [`blameless-post-mortem`](../../../skills/blameless-post-mortem/SKILL.md) — post-mortem + near-miss skill
+- [`root-cause-frameworks`](../../../skills/root-cause-frameworks/SKILL.md) — RCA engine
+- [`premortem`](../../../skills/premortem/SKILL.md) — forward-looking imagined-failure skill
+- [`decision-review`](../../../skills/decision-review/SKILL.md) — past-decision audit skill
+- [`incident-commander`](../../../skills/incident-commander/SKILL.md) — live-incident coordination skill

package/dist/agent-src/commands/brand/identity.md

@@ -0,0 +1,27 @@
+---
+model_tier: high
+name: brand-identity
+pack: brand
+tier: 2
+visibility: internal
+cluster: brand
+sub: identity
+skills: [brand-identity]
+description: Define the brand identity — logo direction, colour story, type story, imagery direction — and the token constraints downstream generation consumes.
+suggestion:
+  eligible: true
+  trigger_description: "define our brand identity, logo and colour direction, type story, visual identity system"
+  trigger_context: "user has a brand strategy and wants the identity system (logo/colour/type/imagery)"
+workspaces:
+  - small-business
+packs:
+  - brand
+---
+
+# /brand:identity
+
+Run the [`brand-identity`](../../skills/brand-identity/SKILL.md) skill — logo
+direction, colour story, type story, imagery direction. brand-identity
+**defines** the tokens/constraints; pack-ai-image's brand-asset generation
+**generates** the marks from them (B → A). Args: `"<strategy ref or brief>"`.
+Run `/brand:strategy` first when no positioning exists.

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

@@ -0,0 +1,27 @@
+---
+model_tier: high
+name: brand-review
+pack: brand
+tier: 2
+visibility: internal
+cluster: brand
+sub: review
+skills: [brand-audit]
+description: Audit emitted UI, copy, and assets against the active brand tokens and voice profile — flag any value not traceable to a brand token or voice rule.
+suggestion:
+  eligible: true
+  trigger_description: "audit our brand consistency, is this on-brand, check this UI/copy against the brand, brand review"
+  trigger_context: "user has emitted UI/copy/assets and wants an on-brand consistency check against tokens + voice"
+workspaces:
+  - small-business
+packs:
+  - brand
+---
+
+# /brand:review
+
+Run the [`brand-audit`](../../skills/brand-audit/SKILL.md) skill — audit emitted
+UI / copy / assets against the active brand tokens (from `state.ui_design` / the
+consumer's `.tokens.json`) and the voice profile, per the `brand-consistency`
+gate. A value not traceable to a brand token or voice rule is flagged off-brand.
+Args: `"<artifact path or description>"`.

package/dist/agent-src/commands/brand/strategy.md

@@ -0,0 +1,27 @@
+---
+model_tier: high
+name: brand-strategy
+pack: brand
+tier: 2
+visibility: internal
+cluster: brand
+sub: strategy
+skills: [brand-strategy]
+description: Define brand positioning, archetype, voice, tone, and messaging over the brand-grounding corpus — the strategy that bounds identity and UI.
+suggestion:
+  eligible: true
+  trigger_description: "define our brand strategy, what's our brand archetype, positioning and messaging, brand voice direction"
+  trigger_context: "user wants brand positioning/archetype/voice/messaging before identity or UI work"
+workspaces:
+  - small-business
+packs:
+  - brand
+---
+
+# /brand:strategy
+
+Run the [`brand-strategy`](../../skills/brand-strategy/SKILL.md) skill —
+positioning, archetype (the 12), voice, tone, messaging frameworks grounded in
+the brand corpus. Output: the strategic frame identity/tokens/voice build on.
+Args: `"<brief>"` (company / product + market). Consumer brand tokens, if
+present, are the source of truth; the corpus fills gaps.

package/dist/agent-src/commands/brand/tokens.md

@@ -0,0 +1,28 @@
+---
+model_tier: high
+name: brand-tokens
+pack: brand
+tier: 2
+visibility: internal
+cluster: brand
+sub: tokens
+skills: [brand-to-tokens]
+description: Derive a DTCG .tokens.json source of truth from brand decisions, then emit CSS vars + Tailwind via the no-Node token generator.
+suggestion:
+  eligible: true
+  trigger_description: "derive brand tokens, turn our brand into design tokens, generate a .tokens.json from the brand"
+  trigger_context: "user has brand identity decisions and wants a DTCG token source of truth + CSS/Tailwind output"
+workspaces:
+  - small-business
+packs:
+  - brand
+---
+
+# /brand:tokens
+
+Run the [`brand-to-tokens`](../../skills/brand-to-tokens/SKILL.md) skill — brand
+decisions → a DTCG `.tokens.json` source of truth → `design-tokens` emits CSS
+vars + Tailwind (no Node). The same `.tokens.json` is the export pack-ai-image's
+brand-asset generation and the greenfield scaffold seed consume; stays on
+`$value`/`$type` so Tokens Studio / Style Dictionary round-trip. Args:
+`"<identity ref>"`.

package/dist/agent-src/commands/brand/voice.md

@@ -0,0 +1,27 @@
+---
+model_tier: high
+name: brand-voice
+pack: brand
+tier: 2
+visibility: internal
+cluster: brand
+sub: voice
+skills: [voice-and-tone-design]
+description: Define the brand voice-and-tone profile — register, do/don't lexicon, and tone shifts by context — the profile the brand-consistency gate checks copy against.
+suggestion:
+  eligible: true
+  trigger_description: "define our brand voice, voice and tone guide, how should our copy sound, tone of voice profile"
+  trigger_context: "user wants a documented voice-and-tone profile for the brand to govern copy"
+workspaces:
+  - small-business
+packs:
+  - brand
+---
+
+# /brand:voice
+
+Run the [`voice-and-tone-design`](../../skills/voice-and-tone-design/SKILL.md)
+skill — register, do/don't lexicon, per-context tone shifts. The voice profile
+is what `/brand:review` and the `brand-consistency` gate check emitted copy
+against. Args: `"<brand/strategy ref>"`. Run `/brand:strategy` first when the
+archetype/positioning is not yet set.

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

@@ -0,0 +1,58 @@
+---
+model_tier: inherit
+name: brand
+pack: brand
+tier: 2
+visibility: internal
+cluster: brand
+type: orchestrator
+description: Brand-as-UX orchestrator — strategy, identity, tokens, review, voice. Routes to the brand-grounding skills that constrain the design layer.
+suggestion:
+  eligible: true
+  trigger_description: "work on our brand, define brand strategy/identity/voice, derive brand tokens, audit brand consistency"
+  trigger_context: "user wants to define or govern a brand layer (positioning, identity, tokens, voice) that constrains the UI"
+workspaces:
+  - small-business
+packs:
+  - brand
+---
+
+# /brand
+
+Top-level orchestrator for the `/brand:*` family — brand as a first-class UX
+layer that **constrains** the UI. Consulted **before**
+[`design-intelligence`](../../skills/design-intelligence/SKILL.md) so brand
+bounds style selection. Designer flow: discover → strategy → identity → tokens →
+application → governance. Derived brand tokens are the export pack-ai-image's
+brand-asset generation and the greenfield scaffold seed consume (B → A).
+
+## Sub-commands
+
+| Sub-command | Routes to | Purpose |
+|---|---|---|
+| `/brand:strategy "<brief>"` | `commands/brand/strategy.md` | Positioning, archetype, voice, tone, messaging over the brand corpus |
+| `/brand:identity "<strategy ref>"` | `commands/brand/identity.md` | Logo/colour/type/imagery direction; defines the token constraints |
+| `/brand:tokens "<identity ref>"` | `commands/brand/tokens.md` | Derive a DTCG `.tokens.json` source of truth → CSS vars + Tailwind |
+| `/brand:review "<artifact>"` | `commands/brand/review.md` | Audit emitted UI/copy/assets against active brand tokens + voice |
+| `/brand:voice "<brand ref>"` | `commands/brand/voice.md` | Define the voice-and-tone profile the consistency gate checks copy against |
+
+## Dispatch
+
+1. Parse `/brand <sub-command> [args]`. Sub-command = first token; match against
+   the table's exact names only.
+2. Look up the sub-command and execute its file verbatim with the remaining args.
+3. Unknown / missing sub-command → print the table and ask which stage to run.
+
+## Rules
+
+- **Do NOT commit, push, or open a PR** — subcommands never do this.
+- **Do NOT chain subcommands.** One `/brand <sub>` per turn.
+- **Consumer brand tokens are the source of truth.** Corpus only fills gaps; a
+  value not traceable to a brand token or voice rule is off-brand (per
+  [`brand-consistency`](../rules/brand-consistency.md)).
+- **Edit `src/` only.** Generated mirrors regenerate.
+
+## See also
+
+- [`brand-strategy`](../../skills/brand-strategy/SKILL.md) · [`brand-identity`](../../skills/brand-identity/SKILL.md) · [`brand-to-tokens`](../../skills/brand-to-tokens/SKILL.md) · [`brand-audit`](../../skills/brand-audit/SKILL.md) · [`voice-and-tone-design`](../../skills/voice-and-tone-design/SKILL.md) — the skills these commands invoke.
+- [`docs/contracts/command-clusters.md`](../docs/contracts/command-clusters.md) — `brand` cluster registration.

package/dist/agent-src/commands/check-current-md.md

@@ -37,7 +37,7 @@ Stop here.
 ### 2. Run the checker
 
 ```bash
-python3 scripts/check_md_language.py <path> [<path> …] --format json
+./scripts-run src/scripts/check_md_language <path> [<path> …] --format json
 ```
 
 Exit codes:
@@ -105,7 +105,7 @@ Edit only the source-of-truth file:
 After all fixes, re-run:
 
 ```bash
-python3 scripts/check_md_language.py <path> [<path> …]
+./scripts-run src/scripts/check_md_language <path> [<path> …]
 ```
 
 ### 6. Mark hashes (only if `.agent-src.uncondensed/` was edited)
@@ -113,7 +113,7 @@ python3 scripts/check_md_language.py <path> [<path> …]
 For each modified source file:
 
 ```bash
-python3 scripts/condense.py --mark-done "{relative_path}"
+./scripts-run src/scripts/condense --mark-done "{relative_path}"
 ```
 
 This keeps `.condensation-hashes.json` consistent with the new content.

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

@@ -112,7 +112,7 @@ For each changed `.md` file:
 7. **MANDATORY: Run condensation quality check on this file:**
 
 ```bash
-python3 scripts/check_condensation.py --format text 2>&1 | grep "{path}"
+./scripts-run src/scripts/check_condensation --format text 2>&1 | grep "{path}"
 ```
 
 If the output contains 🔴 (error) for this file: **STOP. Fix the condensed file before continuing.**
@@ -146,7 +146,7 @@ bash scripts/condense.sh --check
 Must show ✅ (hashes in sync).
 
 ```bash
-python3 scripts/check_condensation.py
+./scripts-run src/scripts/check_condensation
 ```
 
 Must show **zero 🔴 errors**. Warnings (🟡) are acceptable.

package/dist/agent-src/commands/council/debate.md

@@ -87,7 +87,7 @@ to abort.
 ### 5. Invoke the CLI
 
 ```bash
-python3 scripts/council_cli.py debate <topic-path> \
+./scripts-run src/scripts/council_cli debate <topic-path> \
   --output agents/runtime/council/sessions/<date>-<slug>/ \
   --confirm \
   --rounds <N> \
@@ -114,7 +114,7 @@ After all rounds complete, render the last round's JSON via
 `/council` render flow:
 
 ```bash
-python3 scripts/council_cli.py render agents/runtime/council/sessions/<dir>/debate-round-<N>.json
+./scripts-run src/scripts/council_cli render agents/runtime/council/sessions/<dir>/debate-round-<N>.json
 ```
 
 The debate mode uses the structured decision-lens template (Karpathy

package/dist/agent-src/commands/council/default.md

@@ -52,8 +52,8 @@ the host translates that to `--depth deep` on the CLI. If multiple
 active artefacts disagree, **deep wins** (max policy). Explicit
 `rounds:N` overrides depth.
 
-The default comes from `ai_council.min_rounds` in
-`.agent-settings.yml` (default `2` so members critique each other at
+The default comes from `defaults.min_rounds` in the resolved
+`.ai-council.yml` (default `2` so members critique each other at
 least once before convergence). **Do NOT ask the user "how many
 rounds?"** when `rounds:N` is unset or `N <= min_rounds` — proceed
 with the settings default. Ask only when the artefact is genuinely
@@ -88,16 +88,40 @@ Neutrality — context-handoff).
 
 ### 2. Check the council is configured + price table fresh
 
-Read `.agent-settings.yml` → `ai_council`:
+Council config lives in **`.ai-council.yml`**, NOT in `.agent-settings.yml`.
+The CLI (`scripts/ai_council/config.py:resolve_config_path`, per
+[ADR-093](../../../docs/decisions/ADR-093-ai-council-config-user-global.md))
+resolves it — first hit wins:
 
-- If `ai_council.enabled` is false → state that and offer to flip it
-  on. Do not flip it autonomously.
-- If no member has `enabled: true` → list the install commands
-  (`./agent-config keys:install-anthropic`, `./agent-config keys:install-openai`)
-  and stop.
-- If a member is enabled but its `*.key` file is missing or has the
-  wrong mode → tell the user which key to install. Do not fall back
-  to env vars. Ever.
+1. Project-local `<project_root>/agents/settings/.ai-council.yml` (if present).
+2. User-global `~/.event4u/agent-config/settings/.ai-council.yml` — default
+   home; one global config serves every project.
+
+Keys are **top-level** (`enabled`, `defaults:`, `members:`). No `ai_council:`
+parent block; the legacy `ai_council.*` block under `.agent-settings.yml` was
+removed in ADR-093 — never read or recommend that path for council config.
+
+```
+NEVER claim the council "needs `.agent-settings.yml` set up" or an
+`ai_council` block. The CLI's `resolve_config_path` is authoritative:
+run `council:estimate` (Step 3) and read its exit code + message before
+making any "not configured" claim. A user-global `.ai-council.yml` with
+`enabled: true` works from EVERY project, including consumer repos that
+have no project-local config.
+```
+
+CLI is the single source of truth for the configured-check — do not hand-parse
+the YAML to second-guess it. Map its outcomes (Step 4 exit codes + the
+**CLI exits 2** cases below):
+
+- `enabled: false` (top-level in resolved `.ai-council.yml`) → CLI exits 2 with
+  `ai_council.enabled is false`. State that, offer to flip it on. Do not flip
+  it autonomously.
+- No member has `enabled: true` (under top-level `members:`) → list install
+  commands (`./agent-config keys:install-anthropic`,
+  `./agent-config keys:install-openai`) and stop.
+- A member is enabled but its `*.key` file is missing or wrong-mode → tell the
+  user which key to install. Do not fall back to env vars. Ever.
 
 Load the price table via `scripts.ai_council.pricing.load_prices()`
 (auto-bootstraps `agents/runtime/.agent-prices.md` from defaults if missing). Run
@@ -172,8 +196,8 @@ Once the user picks `1`, invoke the same arguments with `run` plus
     [--original-ask "<framing sentence>"]
 ```
 
-`--rounds` defaults to `ai_council.min_rounds` from
-`.agent-settings.yml` (or `2` if unset). Pass `--rounds N` only when
+`--rounds` defaults to `defaults.min_rounds` from the resolved
+`.ai-council.yml` (or `2` if unset). Pass `--rounds N` only when
 the user explicitly asked for a different count or a complex
 artefact justifies more depth — do not pass `--rounds 1` to "save
 money" by default; the settings owner already chose `min_rounds`.
@@ -181,15 +205,15 @@ money" by default; the settings owner already chose `min_rounds`.
 `--depth` defaults to `standard`. Set `--depth deep` when the
 active rule, skill, or command declares `council_depth: deep` in
 its frontmatter; the floor becomes
-`max(ai_council.deep_min_rounds, ai_council.min_rounds)` (default
+`max(defaults.deep_min_rounds, defaults.min_rounds)` (default
 `3`). If `--rounds N` is also passed, `--rounds` wins.
 
 The CLI:
 
 - bundles the artefact via `scripts.ai_council.bundler` (redaction +
   size guard — `BundleTooLarge` exits 2 with the byte count),
-- builds members from `.agent-settings.yml` (refusing if
-  `ai_council.enabled` is false or no member is wired up),
+- builds members from the resolved `.ai-council.yml` (refusing if
+  top-level `enabled` is false or no member is wired up),
 - detects project context via `detect_project_context()`,
 - calls `orchestrator.consult(...)` with the `cost_budget` from
   settings,
@@ -253,8 +277,11 @@ council can act on the project directly.
 
 ## Failure modes
 
-- **CLI exits 2, "ai_council.enabled is false"** → tell the user how
-  to flip it on; do not flip it autonomously.
+- **CLI exits 2, "ai_council.enabled is false"** → tell the user to set
+  top-level `enabled: true` in the resolved `.ai-council.yml` (user-global
+  `~/.event4u/agent-config/settings/.ai-council.yml`, or project-local
+  `<project_root>/agents/settings/.ai-council.yml`) — never in
+  `.agent-settings.yml`. Do not flip it autonomously.
 - **CLI exits 2, "no council member has `enabled: true`"** → list the
   install commands (`./agent-config keys:install-anthropic`,
   `./agent-config keys:install-openai`) and stop.

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

@@ -23,7 +23,7 @@ packs:
 ### 1. Run the portability checker
 
 ```bash
-python3 src/scripts/check_portability.py --format json
+./scripts-run src/scripts/check_portability --format json
 ```
 
 ### 2. Parse findings
@@ -88,7 +88,7 @@ Edit files in `.agent-src.uncondensed/`, then run `bash src/scripts/condense.sh
 After all fixes, re-run:
 
 ```bash
-python3 src/scripts/check_portability.py
+./scripts-run src/scripts/check_portability
 ```
 
 ### 6. Mark hashes
@@ -96,7 +96,7 @@ python3 src/scripts/check_portability.py
 For each modified file:
 
 ```bash
-python3 src/scripts/condense.py --mark-done "{relative_path}"
+./scripts-run src/scripts/condense --mark-done "{relative_path}"
 ```
 
 ## Rules

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

@@ -23,7 +23,7 @@ packs:
 ### 1. Run the reference checker
 
 ```bash
-python3 src/scripts/check_references.py --format json
+./scripts-run src/scripts/check_references --format json
 ```
 
 ### 2. Parse findings
@@ -83,7 +83,7 @@ Edit files in `.agent-src.uncondensed/` (source of truth). Regenerate `dist/agen
 After all fixes:
 
 ```bash
-python3 src/scripts/check_references.py
+./scripts-run src/scripts/check_references
 ```
 
 Show final result.
@@ -93,7 +93,7 @@ Show final result.
 For each modified file:
 
 ```bash
-python3 src/scripts/condense.py --mark-done "{relative_path}"
+./scripts-run src/scripts/condense --mark-done "{relative_path}"
 ```
 
 ## Rules

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

@@ -25,22 +25,51 @@ packs:
 
 ### 1. Resolve input
 
-Accept one of four input paths:
+Accept one of five input paths:
 
 1. **Explicit key** — `/implement-ticket PROJ-123`
 2. **Branch detection** — no arg → `git branch --show-current` + regex
    `[A-Z]+-[0-9]+`
 3. **Pasted text** — markdown block under the command
 4. **URL** — `/implement-ticket https://…/browse/PROJ-123`
+5. **Local ticket bundle** — a path to a bundle ticket file,
+   `/implement-ticket agents/tickets/{slug}/T-NNN-….md` (the
+   ticket-bundle artifact from [`emit-tickets`](emit-tickets.md);
+   format: [`ticket-bundle-format`](../docs/contracts/ticket-bundle-format.md)).
 
 For paths 1, 2, 4: run steps 1-3 of the [`jira-ticket`](jira-ticket.md) command
 (extract ID, fetch ticket, scan for Sentry links). For path 3: parse the pasted
-markdown. If no input resolves:
+markdown. For path 5: parse the bundle ticket — its YAML frontmatter +
+Markdown body map straight onto the work-state envelope with **no engine
+schema fork** (`input.data = {id, title, body, acceptance_criteria}`; the
+ticket `acceptance` list → `acceptance_criteria`, the body → `body`). Then
+apply the **bundle-mode guards** in §1a before driving the engine. If no
+input resolves:
 
 ```
 🎫 Which ticket should I implement? Paste a key, URL, or the ticket text.
 ```
 
+### 1a. Bundle-mode guards (path 5 only)
+
+When the input is a local ticket bundle, three guards run **before** the
+engine — they make a `lite`-tier build safe (per
+[`ADR-101`](../../docs/decisions/ADR-101-ticket-bundle-emission.md)):
+
+- **Dependency-driven selection.** Read the bundle `manifest.yml`
+  `dependency_graph`. Only build a ticket whose `blocked_by` are all
+  `done`. If a dependency is open, STOP and surface it — never build
+  out of order.
+- **Pre-build staleness gate.** Run
+  `python3 src/scripts/lint_ticket_buildable.ts` on the bundle. An
+  `adr_refs` SHA drift is a **hard** stop ("re-emit the bundle"); a
+  `source_refs` SHA drift only **warns** and proceeds (split severity).
+- **Boundary guard.** Carry the ticket's `boundaries`
+  (`must_touch` / `may_touch` / `must_not_touch`) into the
+  `apply-plan` and `review-changes` directives (§4). Any changeset file
+  outside `must_touch ∪ may_touch` HALTS with an escalation surface —
+  prose "do not touch" does not bind a cheap agent; this guard does.
+
 ### 2. Prepare the state file
 
 The engine persists everything in a JSON state file (default:
@@ -59,10 +88,11 @@ Three cases, in this order:
   ./agent-config migrate
   ```
 
-  Unified `migrate` writes `.work-state.json`, renames source to
-  `.implement-ticket-state.json.bak`, sweeps other legacy install
-  artefacts (see `docs/contracts/migrate-command.md`). Idempotent and
-  safe to skip if already done. After this, treat the run as **Resume**.
+  The unified `migrate` command writes `.work-state.json`, renames the
+  source to `.implement-ticket-state.json.bak`, and sweeps any other
+  legacy install artefacts in the same pass (see
+  `docs/contracts/migrate-command.md`). Idempotent and safe to skip if
+  already done. After this, treat the run as **Resume**.
 - **Fresh run** — no state file at all. Write the resolved ticket to
   `ticket.json` (id, title, body, acceptance_criteria) and pass it via
   `--ticket-file ticket.json`. Honour `roles.active_role` from

package/dist/agent-src/commands/knowledge/cross-repo.md

@@ -45,7 +45,7 @@ concept (≥ 1 term > 2 chars). A `--path-scope` glob narrows the search and is
 ### 2. Run the retrieval
 
 ```bash
-python3 scripts/cross_repo_retrieve.py "<query>" [--path-scope <glob>] [--max-chunks 8]
+./scripts-run src/scripts/cross_repo_retrieve "<query>" [--path-scope <glob>] [--max-chunks 8]
 ```
 
 The script reads opted-in siblings only, runs a targeted path-glob + content

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

@@ -111,7 +111,7 @@ On confirmation:
 ### 6. Cross-link
 
 If the entry has a `references:` block pointing to a skill, rule, or
-ADR file, run `python3 scripts/check_references.py` to verify the
+ADR file, run `./scripts-run src/scripts/check_references` to verify the
 link resolves. Broken link → block and ask the user to fix.
 
 ## When to reject