@event4u/agent-config 6.1.0 → 7.0.0

package/dist/agent-src/commands/mission/upgrade.md

@@ -0,0 +1,182 @@
+---
+model_tier: medium
+name: mission-upgrade
+pack: engineering-base
+intent: "Upgrade a Laravel application one major version (10→11, 11→12) via the gated /work engine with a structured breaking-change catalog"
+routes_to: [dependency-upgrade]
+replaces: []
+tier: 1
+visibility: advanced
+skills: [dependency-upgrade]
+framework: laravel
+description: Gated Laravel major-version upgrade mission — provisional branch, breaking-change catalog, size-tier surfaced, git-as-rollback. Never auto-commits or auto-PRs.
+suggestion:
+  eligible: true
+  trigger_description: "upgrade Laravel to version X, Laravel 10 to 11 upgrade, bump framework major version"
+  trigger_context: "composer.json present with a laravel/framework constraint and user asks for a major version bump"
+workspaces:
+  - agent-config-maintainer
+packs:
+  - engineering-base
+disable-model-invocation: true
+---
+
+# /mission:upgrade
+
+## Overview
+
+Structured, catalog-driven Laravel major-version upgrade through the existing gated `/work` engine. **Not** a new autonomous runtime:
+
+- Every phase uses `/work`'s confidence-band gating, N=3 retry budget, persisted `.work-state.json`.
+- Rollback = git. Changes land on a provisional branch (`mission/upgrade-…`); revert is `git reset` or `git revert`, never a daemon.
+- No auto-commit, no auto-PR. User reviews before anything is staged.
+
+## Single-mission-per-branch guard
+
+Before starting, confirm:
+
+1. Working tree clean (`git status --porcelain` empty).
+2. No other `mission/*` branch checked out.
+3. No `.work-state.json` with a mission envelope present.
+
+Any fail → surface the conflict and stop. Run:
+
+```bash
+python3 src/scripts/lint_missions.ts --check-precondition upgrade .
+```
+
+(Currently a documented stub — full live-repo check lands in Phase 1 PoC.)
+
+## Inputs
+
+| Input | Required | Description |
+|---|---|---|
+| `target_version` | Yes | Target major version to upgrade to (`11`, `12`, …) |
+| `framework` | No (default: `laravel`) | Framework slug — auto-detected from `composer.json` |
+
+## Execution flow
+
+### 1. Resolve inputs
+
+Accept inline arguments:
+
+```
+/mission:upgrade 11
+/mission:upgrade target_version=11 framework=laravel
+```
+
+`target_version` missing → ask once (numbered options per [`user-interaction`](../../rules/user-interaction.md)).
+
+### 2. Surface the size tier
+
+Read `size_tier` from `src/missions/upgrade/mission.yaml`; display before starting:
+
+```
+> Mission: upgrade Laravel {from} → {target_version}
+> Size tier: standard (touches public API and configuration)
+> Provisional branch: mission/upgrade-laravel-{from}-to-{target_version}
+>
+> 1. Proceed on a new branch
+> 2. Adjust size tier — I'll override
+> 3. Abort
+```
+
+`standard` is the floor for upgrade missions (public API + config changes).
+
+### 3. Create the provisional branch
+
+```bash
+git checkout -b mission/upgrade-laravel-{from}-to-{target_version}
+```
+
+Per [`scope-control`](../../rules/scope-control.md), requires explicit user confirmation (step 2 provides it).
+
+### 4. Load the breaking-change catalog
+
+Read `src/missions/upgrade/laravel-10-to-11.yaml` (or the matching version). Lists breaking changes with `detection`, `fix`, `verification` command blocks restricted to safe prefixes (`composer`, `php`, `php artisan`, `git`, `sed`, `rector`, `vendor/bin/*`).
+
+### 5. Drive the /work engine
+
+```bash
+./agent-config work \
+    --state-file .work-state.json \
+    --prompt-file prompt.txt
+```
+
+Write a structured prompt to `prompt.txt` describing the mission phases (`analyze` → `plan` → `implement` → `test` → `verify` → `report`) referencing the catalog entries the engine should address.
+
+**Candidate phase (opt-in, never auto-on):** the `test` → `verify` step MAY use
+[`verify-repair-loop`](../../skills/verify-repair-loop/SKILL.md) when the upgrade's
+catalog fixes are test-driven — a bounded generate→run→revise→re-run loop gated by
+a numeric threshold (then a judge confirms), reusing the same N=3 cap. Use
+test/quality verdicts only; the live-app Playwright verdict stays deferred. The
+loop is multi-turn reasoning, not a runtime — honors
+[`no-runtime-boundary`](../../docs/contracts/no-runtime-boundary.md) exactly as the
+mission does.
+
+Handle exit codes per the `/work` contract:
+
+| Exit | Meaning | Action |
+|---|---|---|
+| `0` | SUCCESS — final report on stdout | Go to step 6 |
+| `1` | BLOCKED — halt surface on stdout | Inspect `questions[0]`, respond, re-run |
+| `2` | Config/IO error | Surface stderr to the user, stop |
+
+### 6. Rollback path
+
+Fix step fails beyond the N=3 budget:
+
+```bash
+git revert HEAD        # revert the last mission commit
+# or
+git reset --hard HEAD~1  # if no remote tracking the branch yet
+```
+
+Rollback is always manual — the mission surfaces the command, never auto-runs it.
+
+### 7. Final report and close-prompt
+
+On `/work` exit `0`, surface the delivery report unchanged, then append:
+
+```
+> Mission: upgrade complete. Review the changes on branch mission/upgrade-…
+>
+> 1. /commit — stage + commit per the delivery report
+> 2. /create-pr — open a pull request from this branch
+> 3. Keep working — hold the state file for follow-up
+> 4. Discard — delete .work-state.json and reset the branch
+```
+
+Per [`scope-control`](../../rules/scope-control.md), git operations are permission-gated. Never run `/commit` or `/create-pr` without the user choosing them.
+
+## Rules
+
+- Honour [`scope-control`](../../rules/scope-control.md), [`non-destructive-by-default`](../../rules/non-destructive-by-default.md), [`minimal-safe-diff`](../../rules/minimal-safe-diff.md), [`verify-before-complete`](../../rules/verify-before-complete.md).
+- Every catalog `command:` field is restricted to the safe-prefix allowlist. Never run arbitrary shell from the catalog.
+- The N=3 retry budget from [`autonomous-execution`](../../rules/autonomous-execution.md) applies per breaking-change fix step, not per mission.
+- Never skip the size-tier surface step — user must see scope before the provisional branch is created.
+
+## Breaking-change catalog
+
+Knowledge source: `src/missions/upgrade/laravel-10-to-11.yaml`. Versionable, diffable, CI-tested via `lint_missions.py`. Each entry carries:
+
+- `id` — unique checkpoint key (used in `.work-state.json`)
+- `severity` — `critical` / `high` / `medium` / `low`
+- `detection` — how to check if this change affects the project
+- `fix` — how to apply the fix (may reference a Rector rule)
+- `verification` — how to confirm the fix is complete
+
+## See also
+
+- [`/work`](../../work/command.md) — the gated engine this mission drives
+- [`dependency-upgrade`](../../skills/dependency-upgrade/SKILL.md) — the stack-agnostic upgrade skill this mission specialises
+- `src/missions/upgrade/mission.yaml` — the mission manifest
+- `src/missions/upgrade/laravel-10-to-11.yaml` — the breaking-change catalog
+- `src/scripts/lint_missions.ts` — validates the manifest and catalog
+
+## Examples
+
+```
+/mission:upgrade 11
+/mission:upgrade target_version=11
+```

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

@@ -28,7 +28,7 @@ Skill audit: measure, find duplicates/merge candidates, run linter, present find
 ### 1. Measure baseline
 
 ```bash
-total=$(ls -d .agent-src.uncondensed/skills/*/SKILL.md 2>/dev/null | wc -l | tr -d ' ')
+total=$(ls -d src/skills/*/SKILL.md 2>/dev/null | wc -l | tr -d ' ')
 total_lines=$(cat .agent-src.uncondensed/skills/*/SKILL.md 2>/dev/null | wc -l | tr -d ' ')
 echo "Skills: $total, Total lines: $total_lines"
 
@@ -43,7 +43,7 @@ done | sort -rn | head -15
 ### 2. Run skill linter
 
 ```bash
-python3 scripts/skill_linter.py --pairs --duplicates .agent-src.uncondensed/skills/*/SKILL.md 2>&1 | tail -20
+./scripts-run src/scripts/skill_linter --pairs --duplicates src/skills/*/SKILL.md 2>&1 | tail -20
 ```
 
 Report FAIL/WARN counts. Do NOT fix linter issues here — that's the linter's or `skill-reviewer`'s job.

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

@@ -44,7 +44,7 @@ side; this command reads the same shape and dispatches each step.
 Run the linter against the resolved file:
 
 ```
-python3 scripts/lint_orchestration_dsl.py --file <path>
+./scripts-run src/scripts/lint_orchestration_dsl --file <path>
 ```
 
 Exit code ≠ 0 → surface the linter output and stop. **Never** run

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

@@ -74,12 +74,14 @@ Run, in order:
 | Behind | Overlapping open PR touches same files? | Action |
 |---|---|---|
 | `0` | — | Proceed to Step 2 |
-| `1–N` | No | Surface the count, ask: rebase / merge-main / proceed-anyway / cancel |
+| `1–N` | No | **Merge `origin/main` in** — `git merge origin/{target-base} --no-edit` — then proceed. No need to ask; state that you did it. Abort + surface only on conflict. |
 | `1–N` | **Yes** | STOP — surface the overlapping PR number, ask: stack on top of it / wait for it to land / proceed-anyway-and-accept-conflicts / cancel |
 
-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.
+A branch behind its base is **not** current — bring `main` in **before** the PR
+exists, never after. Auto-merge when there is no file overlap (the common case);
+only ask when an overlapping open PR makes the base genuinely ambiguous. Never
+improvise the base or silently proceed when behind **and** overlapping. The
+10-second fetch beats hours of rebase reconciliation after the parent PR lands.
 
 ### 1c. PR-gate — archive completed roadmaps (MANDATORY)
 

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

@@ -146,6 +146,13 @@ Produce one combined report:
   ask before fixing 🟡 findings, report 🟢 as suggestions
 - If all five returned `apply` → the diff is ready; report and stop
 
+**Opt-in (never auto-on):** when the `revise` findings are *test-driven*
+(failing checks, not subjective craft) and you want bounded auto-repair before
+re-review, hand off to
+[`verify-repair-loop`](../skills/verify-repair-loop/SKILL.md) — a
+generate→run→revise→re-run loop gated by a numeric threshold, then a judge
+confirms. Surface as a single follow-up line; do not auto-invoke.
+
 ### 7. Quality tools (verbosity-gated)
 
 Per `verbosity.routine_confirmations` (default `false`):
@@ -201,6 +208,7 @@ Per `verbosity.routine_confirmations` (default `false`):
 - [`/prepare-for-review`](prepare-for-review.md) — updates `main` and merges the full base-branch chain into the target branch (used by step 1)
 - [`subagent-orchestration`](../skills/subagent-orchestration/SKILL.md) — dispatch and model-pairing rules
 - [`/do-and-judge`](do-and-judge.md) — implementer + judge loop for a single change
+- [`verify-repair-loop`](../skills/verify-repair-loop/SKILL.md) — opt-in test-verdict-gated iterate-to-green loop (step 6 hand-off); judge confirms after the numeric gate
 - [`/judge`](judge.md) — standalone judge, no review-changes dispatch
 - [`code-review`](../skills/code-review/SKILL.md) — human-oriented review patterns (tone, feedback handling)
 - [`role-contracts`](../docs/guidelines/agent-infra/role-contracts.md#reviewer) — Reviewer mode output contract (Summary / Risks / Findings / Required actions / Verdict)

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

@@ -0,0 +1,73 @@
+---
+model_tier: high
+name: roadmap-materialize
+pack: product-basic
+tier: 2
+visibility: internal
+cluster: roadmap
+sub: materialize
+skills: [emit-tickets]
+description: Materialise a roadmap into a self-contained, importable ticket bundle under agents/tickets/
+suggestion:
+  eligible: true
+  trigger_description: "turn this roadmap into tickets, materialise the roadmap, mach Tickets aus der Roadmap"
+  trigger_context: "an existing agents/roadmaps/*.md the user wants decomposed into build-ready tickets"
+workspaces:
+  - agent-config-maintainer
+packs:
+  - meta
+---
+
+# /roadmap materialize
+
+## Instructions
+
+A thin wrapper that delegates to the [`emit-tickets`](../../../skills/emit-tickets/SKILL.md)
+skill. This command **orchestrates**; the skill owns the procedure.
+
+### 1. Resolve the roadmap
+
+Accept a roadmap path argument (`/roadmap:materialize agents/roadmaps/{roadmap}.md`)
+or, with no argument, ask which roadmap to materialise. The roadmap must already
+exist (authored via `/roadmap:create` + `roadmap-writing`); this command does
+**not** author roadmaps.
+
+### 2. Run the skill
+
+Invoke [`emit-tickets`](../../../skills/emit-tickets/SKILL.md). It reads the
+roadmap, writes one ticket per materialisable step into
+`agents/tickets/{slug}/`, sets each ticket's `model_tier`, writes `manifest.yml`
+(acyclic dependency graph + empty `linear_state`), writes
+`<!-- ticket: T-NNN -->` markers back into the roadmap, and regenerates
+`agents/tickets/_registry.yml`. Format contract:
+[`ticket-bundle-format`](../../../../docs/contracts/ticket-bundle-format.md).
+
+### 3. Gate on buildability
+
+Run the build-readiness lint before handing back:
+
+```bash
+python3 src/scripts/lint_ticket_buildable.py
+```
+
+A `lite` ticket that fails the self-containedness floor is rejected or escalated
+to `medium` — never shipped under-specified.
+
+### 4. Hand back — HARD STOP
+
+```
+BUNDLE MATERIALISED → STOP. NEVER AUTO-CREATE TRACKER ISSUES OR EXECUTE.
+MATERIALISE = ARTIFACT ONLY. PASTE/MCP/BUILD NEED A FRESH USER VERB.
+```
+
+Emit a single hand-back line citing the bundle path + the ticket count. The user
+pastes a ticket into Linear/Jira (or creates it via MCP), or runs a per-ticket build
+(`/implement-ticket <bundle>/T-NNN.md`) on a later turn, explicitly.
+
+### Rules
+
+- **Do NOT author the roadmap** — this consumes an existing one.
+- **Do NOT auto-create tracker issues** — paste/MCP is a separate user-invoked step (ADR-102, no API export).
+- **Do NOT commit or push.**
+- Bundle is the source of truth; the tracker is a projection
+  ([`ticket-bundle-format`](../../../../docs/contracts/ticket-bundle-format.md)).

package/dist/agent-src/commands/skill/preview.md

@@ -40,7 +40,7 @@ argument. Missing name → print usage and stop.
 ### 2. Run the previewer
 
 ```bash
-python3 scripts/skill_preview.py <name>
+./scripts-run src/scripts/skill_preview <name>
 ```
 
 Add `--technical` for the raw frontmatter + numbered step list; default is the

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

@@ -43,7 +43,7 @@ neither resolves, it prints the available roles and stops — do not guess.
 ### 2. Run the recommender
 
 ```bash
-python3 scripts/skill_discovery.py --role <role>
+./scripts-run src/scripts/skill_discovery --role <role>
 ```
 
 Optional flags: `--format json` (machine-readable), `--limit N` (results per

package/dist/agent-src/commands/threat-model.md

@@ -79,7 +79,7 @@ Produce one combined report with these sections, in order:
 5. **Required controls** — the minimum set of validations, authorization
    checks, logging, and negative tests that must exist before the change ships
 6. **Open questions** — anything the skills flagged as uncertain and the user
-   must answer before impl starts
+   must answer before implementation starts
 
 ### 5. Decide next step
 
@@ -91,7 +91,7 @@ After the report, ask:
 > 3. Stop here — threat model is the deliverable, no implementation yet
 ```
 
-- On **1**: hand off to the impl flow (e.g., `feature-plan`,
+- On **1**: hand off to the implementation flow (e.g., `feature-plan`,
   `bug-fix`, or direct edit) with the required-controls list pinned
 - On **2**: re-gather context and re-dispatch
 - On **3**: save the report as the deliverable, stop
@@ -114,7 +114,7 @@ After the report, ask:
 - NEVER merge `threat-modeling` and `authz-review` outputs into a single
   block — each skill owns its format
 - NEVER write production code in the same turn as this command — the
-  deliverable is the report; impl is a separate step
+  deliverable is the report; implementation is a separate step
 - NEVER mark the change "safe" if any 🔴 abuse case has no control
 
 ## See also
@@ -124,4 +124,4 @@ After the report, ask:
 - [`data-flow-mapper`](../skills/data-flow-mapper/SKILL.md) — trace specific data through the change
 - [`blast-radius-analyzer`](../skills/blast-radius-analyzer/SKILL.md) — enumerate affected call sites
 - [`security-sensitive-stop`](../rules/security-sensitive-stop.md) — the trigger rule
-- [`minimal-safe-diff`](../rules/minimal-safe-diff.md) — keep the impl scoped
+- [`minimal-safe-diff`](../rules/minimal-safe-diff.md) — keep the implementation scoped

package/dist/agent-src/commands/upstream-contribute.md

@@ -126,9 +126,9 @@ Frontmatter must have `source: package`.
 ### 6. Run quality gates
 
 ```bash
-python3 scripts/skill_linter.py --all           # 0 FAIL required
-python3 scripts/check_condensation.py            # No 🔴 errors for this file
-python3 scripts/condense.py --generate-tools    # Regenerate symlinks
+./scripts-run src/scripts/skill_linter --all           # 0 FAIL required
+./scripts-run src/scripts/check_condensation            # No 🔴 errors for this file
+./scripts-run src/scripts/condense --generate-tools    # Regenerate symlinks
 bash scripts/condense.sh --check                # dist/agent-src/ in sync with source
 bash scripts/condense.sh --check-hashes         # All hashes match
 ```

package/dist/agent-src/commands/video/from-script.md

@@ -57,7 +57,7 @@ call has happened yet.
 
 ### 2. Load config + resolve providers
 
-Source `scripts/ai-video/lib/load-config.sh`. Resolve image / video
+Source `scripts/media/lib/load-config.sh`. Resolve image / video
 provider in this order: command flag → `agents/.ai-video.xml` default
 → fail with the available-providers list.
 
@@ -156,4 +156,4 @@ cost (commit) or summed modeled `cost_estimate` labeled *modeled*
 - [`/video:scene`](scene.md) — single-scene iteration
 - [`/video:storyboard`](storyboard.md) — image-only contact sheet
 - [`/video:stitch`](stitch.md) — re-stitch after operator edits
-- [`scripts/ai-video/lib/adapter-contract.md`](../../../scripts/ai-video/lib/adapter-contract.md)
+- [`scripts/media/lib/adapter-contract.md`](../../../scripts/media/lib/adapter-contract.md)

package/dist/agent-src/commands/video/from-song.md

@@ -81,7 +81,7 @@ absent from `dist/agent-src/personas/` or `dist/agent-src/skills/`. No network
 call has happened yet.
 
 Then confirm the **runtime helper scripts** exist and are executable —
-`scripts/ai-video/lib/probe-audio.sh`, `scripts/ai-video/lib/load-config.sh`,
+`scripts/ai-video/lib/probe-audio.sh`, `scripts/media/lib/load-config.sh`,
 `scripts/ai-video/stitch.sh` — so a missing script fails here (with the
 path), not mid-run at Step 2/9.
 
@@ -156,7 +156,7 @@ if they really mean it; the guard never silently proceeds.
 
 ### 4. Load config + resolve providers
 
-Source `scripts/ai-video/lib/load-config.sh`. Resolve image / video
+Source `scripts/media/lib/load-config.sh`. Resolve image / video
 provider: command flag → `agents/.ai-video.xml` default → fail with the
 available-providers list. A **malformed XML** or a default/flag naming a
 provider with no `scripts/ai-video/adapters/<id>.sh` → fail fast here
@@ -488,4 +488,4 @@ labeled *modeled* (preview). No commit. No push.
 - [`/video:stitch`](stitch.md) — re-stitch after operator edits
 - [`song-to-script`](../../skills/song-to-script/SKILL.md) — audio →
   timed scene script
-- [`scripts/ai-video/lib/adapter-contract.md`](../../../scripts/ai-video/lib/adapter-contract.md)
+- [`scripts/media/lib/adapter-contract.md`](../../../scripts/media/lib/adapter-contract.md)

package/dist/agent-src/commands/video/scene.md

@@ -53,7 +53,7 @@ scripts/ai-video/lib/validate-deps.sh .agent-src.uncondensed/commands/video/scen
 
 ### 2. Load config + resolve providers
 
-Source `scripts/ai-video/lib/load-config.sh`. Resolve image / video
+Source `scripts/media/lib/load-config.sh`. Resolve image / video
 provider: flag → `agents/.ai-video.xml` default → fail with the
 available-providers list.
 

package/dist/agent-src/commands/video/storyboard.md

@@ -42,7 +42,7 @@ scripts/ai-video/lib/validate-deps.sh .agent-src.uncondensed/commands/video/stor
 
 ### 2. Load config + resolve image provider
 
-Source `scripts/ai-video/lib/load-config.sh`. Resolve image provider:
+Source `scripts/media/lib/load-config.sh`. Resolve image provider:
 `--image-provider` flag → `agents/.ai-video.xml` default → fail. Reject
 `--video-provider` (this command produces no clips).
 

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

@@ -24,7 +24,7 @@ video creation. Reads provider keys + defaults from
 [`agents/.ai-video.xml`](../../agents/templates/.ai-video.xml.example) (gitignored
 real file; example shipped). Every subcommand is **dry-run by default**;
 network calls require explicit per-turn confirmation per the adapter
-contract under [`scripts/ai-video/lib/adapter-contract.md`](../../scripts/ai-video/lib/adapter-contract.md).
+contract under [`scripts/media/lib/adapter-contract.md`](../../scripts/media/lib/adapter-contract.md).
 
 ## Sub-commands
 
@@ -71,5 +71,5 @@ contract under [`scripts/ai-video/lib/adapter-contract.md`](../../scripts/ai-vid
 
 ## See also
 
-- [`scripts/ai-video/lib/adapter-contract.md`](../../scripts/ai-video/lib/adapter-contract.md) — provider adapter v1 contract
-- [`docs/contracts/command-clusters.md`](../../docs/contracts/command-clusters.md) — `video` cluster registration
+- [`scripts/media/lib/adapter-contract.md`](../../scripts/media/lib/adapter-contract.md) — provider adapter v1 contract
+- [`docs/contracts/command-clusters.md`](../docs/contracts/command-clusters.md) — `video` cluster registration

package/dist/agent-src/contexts/communication/rules-auto/source-of-truth-mechanics.md

@@ -66,9 +66,9 @@ disable-model-invocation: true
 **Creating a new command:**
 
 1. Create `src/agent-src/commands/{name}.md` (use template)
-2. Run `python3 src/scripts/skill_linter.py` — must be 0 FAIL
+2. Run `./scripts-run src/scripts/skill_linter` — must be 0 FAIL
 3. Condense via `/condense`, which writes to `dist/agent-src/commands/`
-4. Run `python3 src/scripts/condense.py --generate-tools` — creates Claude symlink automatically
+4. Run `./scripts-run src/scripts/condense --generate-tools` — creates Claude symlink automatically
 
 **Never** create `.claude/skills/{name}/SKILL.md` manually for commands — always use the symlink workflow.
 
@@ -92,7 +92,7 @@ Commands have `disable-model-invocation: true` in their frontmatter.
 | Edit existing file | Edit in `src/`, condense to `dist/agent-src/` |
 | Create new `.md` | Create in `src/`, condense to `dist/agent-src/` |
 | Create new non-`.md` | Create in `src/`, run `bash src/scripts/condense.sh --sync` |
-| Create new command | Create in `src/agent-src/commands/`, sync, `python3 src/scripts/condense.py --generate-tools` |
+| Create new command | Create in `src/agent-src/commands/`, sync, `./scripts-run src/scripts/condense --generate-tools` |
 | Delete a file | Delete from `src/` and `dist/agent-src/` |
 | Check what needs condensation | `bash src/scripts/condense.sh --changed` |
 | Mark file as condensed | `bash src/scripts/condense.sh --mark-done {path}` |

package/dist/agent-src/contexts/communication/rules-auto/user-interaction-mechanics.md

@@ -82,7 +82,7 @@ position or framing:
    underneath.
 
 Mechanical backstop:
-`python3 scripts/check_reply_consistency.py --stdin < draft.md`
+`./scripts-run src/scripts/check_reply_consistency --stdin < draft.md`
 (non-zero exit on any rule above). Self-scan is the primary gate;
 the script is the deterministic safety net for ambiguous cases.