chore: sync private v5.1.0 (fdc9b54)

.cursor-plugin/plugin.json

@@ -15,7 +15,7 @@
     "anti-slop"
   ],
   "license": "PolyForm-Noncommercial-1.0.0",
-  "version": "5.0.0",
+  "version": "5.1.0",
   "homepage": "https://github.com/agent-kit-startup/agent-kit",
   "repository": "https://github.com/agent-kit-startup/agent-kit",
   "logo": "dashboard/logo.svg",

.cursor/agent-kit.json

@@ -1,6 +1,6 @@
 {
   "schemaVersion": 1,
-  "version": "5.0.0",
+  "version": "5.1.0",
   "protected": [
     ".cursor/HANDOFF.md",
     ".cursor/agents/test-suites.md",

.cursor/agents/mission-kit-comms.md

@@ -0,0 +1,35 @@
+---
+name: mission-kit-comms
+description: Dedicated Mission Kit comms/community agent. Drafts recap, release, contributor-ask, and reply copy under explicit HITL gates. Not an Agent Persona (personas are chat chrome only).
+model: claude-sonnet-4
+readonly: false
+---
+
+# Mission Kit comms agent
+
+You draft adoption and contributor copy for **Mission Kit**. You do not post.
+
+Follow skill `registry/skills/community/mission-kit-comms/SKILL.md` (factory overlay: `.cursor/skills/community/mission-kit-comms/SKILL.md`). Tone: short blocks, one ask per message ([ux-message-flows](../skills/community/ux-message-flows/SKILL.md)). Prompt shape: [prompts-markdown](../skills/community/prompts-markdown/SKILL.md).
+
+## HITL contract
+
+- Before any **public post or public reply**, Ask questions with `Post this` / `Edit draft` / `Discard` (chat numbered list if the tool is missing).
+- Skip or cancel = stop. Never silent cross-network posting.
+- Never `/git-prod`. Never buy ads. Never submit Cursor Marketplace (parked plan owns that).
+- Never load Agent Personas (`autopilot` / `night-shift` / `ghost-runner`) as the poster.
+
+## What you may do
+
+- Draft from [docs/comms-templates/](../../docs/comms-templates/recap.md) and the [calendar](../../docs/comms-content-calendar.md).
+- Run `node .cursor/scripts/comms-draft.mjs` for local files under `.cursor/comms-drafts/`.
+- Align contributor asks with [CONTRIBUTING](../../docs/CONTRIBUTING.md), [contribute-upstream](../../docs/contribute-upstream.md), CoC, SUPPORT, SECURITY.
+
+## What you must not do
+
+- HTTP post, webhook, or `gh` issue comment that is marketing without Ask.
+- Put tokens, webhooks, or cookies in files that can be committed.
+- Claim product behavior that is not in CHANGELOG 5.0.0 / getting-started / five-layer matrix.
+
+## Naming
+
+Mission Kit = product. Agent Kit = CLI/npm/slash/pack. Mission Control = dashboard.

.cursor/commands/git-prod.md

@@ -19,7 +19,11 @@ Follow the **git prod** routine to promote `origin/staging` to `origin/main` (pr
    **Advisory (does not replace confirmation):** if Blocking untriaged `.cursor/memory/plan-monitor-*.md` match themes in the staging→main delta, mention them once in the summary. Never steal this Ask; Field Report / `/plan-review-triage` stay attention/HITL SoT.
 
    **Fallback:** if Ask questions tool unavailable, ask for explicit confirmation in chat.
-5. Run merge to main, push main, create/push annotated vX.Y.Z tag (when absent), confirm production.
+5. Run merge to main, then push with the authorized inline form only:
+
+   `ALLOW_MAIN_PUSH=1 git push origin main`
+
+   Bare `git push origin main` stays denied by `agent-kit guard shell` and `git-hooks/pre-push`. Do not export `ALLOW_MAIN_PUSH=1` as a session environment variable. Do not add `--force`, `--no-verify`, or a non-main destination. Then create/push annotated vX.Y.Z tag (when absent) and confirm production. Details: `autogit/gitupdate.md` Prompt git prod step 9.
 6. Update `.cursor/HANDOFF.md` ("promoted to production") and memory-loop WRITE if it applies.
 7. In this monorepo: annotated tags trigger `publish-npm` + `sync-public` CI; `pnpm git:trigger-public-sync` fallback when needed per `autogit/gitupdate.md`.
 8. **Post-prod verification (mandatory in this monorepo):** before ending, check tag CI jobs (`publish-npm`, `sync-public`), `npm view @dadado/agent-kit-cli version`, **public sync PR merged** (not CI-green alone), public `main` sync commit, and public GitHub Release Latest. Report each row. Silent npm success with a stale public Releases badge, or CI-green with an unmerged sync PR, is a kit failure mode; see `autogit/gitupdate.md` step 12.5.

.cursor/commands/git-staging.md

@@ -10,7 +10,7 @@ Follow the **git staging** routine to bring local changes to the pre-production
 **Runs in the main window by default.** Do not dispatch this command to a Task subagent by default; Task isolation is opt-in and used only when the kit repo wants it.
 
 1. **Read** the "Prompt: git staging" section in `autogit/gitupdate.md` (when it exists).
-2. **Staging hygiene (monitors):** if `git status` shows untracked or unrelated dirty `.cursor/memory/plan-monitor-*.md`, **warn** before commit. Stage memory/monitor files **add-by-name only**; never broad `git add` of `.cursor/memory/` WIP into a product commit (ADR `decisions/2026-07-27_plan-monitor-consumer-awareness.md`, external-review staging hygiene).
+2. **Staging hygiene (monitors):** if `git status` shows untracked or unrelated dirty `.cursor/memory/plan-monitor-*.md`, **warn** before commit. Stage memory/monitor files **add-by-name only**; never broad `git add` of `.cursor/memory/` WIP into a product commit (ADR `decisions/2026-07-27_plan-monitor-consumer-awareness.md`, external-review staging hygiene). Warn/add-by-name forbids sweeps; it does **not** mean leave soft kit artifacts dirty. Inventory remaining safe dirt into a `docs(memory):` / `chore(kit):` bucket and ship it in this run (SoT: `autogit/gitupdate.md` Prompt git staging §1 and §7). Do not stop-and-quiz solely because a path looks out of the current flow.
 3. **Lint evidence (required when code/format paths change):** before claiming staging-ready, **run** the repo formatter/linter on touched files and **record the command + result** (pass/fail) in the worker summary or tick notes. Writing `Staging ready: yes` or the contract string alone is **not** evidence. Pure markdown / docs-only with no applicable linter: state `none applicable`. Same gate as `/run-plan` Staging-ready lint gate. **Dashboard CSS/HTML only** (`dashboard/dashboard.html` and similar, outside Biome scope): record `Tests: none applicable (dashboard-CSS); covered by plugin-ux-validation` when the UX suite pins the change (ADR `decisions/2026-07-29_dashboard-css-lint-evidence-convention.md`); do not claim Biome covered the HTML.
 4. Run in order: validation (not on `main`), CHANGELOG (`[Unreleased]`), checkout staging, pull, working branch, Conventional Commits, push, MR/PR (**always `--base staging` / target `staging`**), merge, cleanup.
 5. **Evidence-checks gate (before merge / Gaps-none):** run `gh pr checks <N>` and confirm `build` (including the **Evidence checks** step) is green. If Evidence checks fail, regenerate and re-push before merge; do not write HANDOFF `- **Gaps:** none` over red. ADR: `2026-08-01_evidence-checks-merge-gate`.

.cursor/commands/handoff.md

@@ -40,9 +40,9 @@ Update the handoff document to preserve current state and allow continuation in
    If they pick automatic, save `{ "autoHandoff": true }` in `.cursor/context/config.json`.
 
 4. **DevOps spine (suggest, do not run without being asked):**
-   - If the phase produced commitable code, suggest `/git-staging` to promote to pre-prod.
+   - If the phase produced commitable changes (including versioned HANDOFF, `plan-monitor-*.md`, or `_index.md` Audits rows), suggest `/git-staging`. That command follows `autogit/gitupdate.md` inventory → theme-bucket → ship. Do not treat HANDOFF/memory as skippable dirt.
    - If there was an error or a tradeoff decision, suggest a memory-loop WRITE (`.cursor/memory/`).
-   - Never suggest committing directly to `main`; production only via `/git-prod` after staging.
+   - Never suggest committing directly to `main`; production only via `/git-prod` after staging. `/git-prod` still requires a clean tree except hard excludes; staging must leave it that way.
 
 5. **Respond to the user:**
    > "Handoff updated! Continue: `/continue-plan`. With code ready: `/git-staging`. Production: `/git-prod`."

.cursor/commands/plan-external-review.md

@@ -7,7 +7,7 @@ description: Arm an optional external plan audit after /run-plan exhausts its im
 
 ## Goal
 
-Manually arm **optional plan audits** (external plan review via Claude Code) after `/run-plan` has exhausted implementable to-dos. Claude writes an evidence-based monitor under `.cursor/memory/plan-monitor-*.md`. Cursor triage of findings is a **later** step (not this command).
+Manually arm **optional plan audits** (external plan review via Claude Code or Cursor Agent) after `/run-plan` has exhausted implementable to-dos. The reviewer writes an evidence-based monitor under `.cursor/memory/plan-monitor-*.md`. Cursor triage of findings is a **later** step (not this command).
 
 ## When to Use
 
@@ -34,7 +34,7 @@ If any are missing: stop. Do **not** claim a review ran. Tell the user to run `a
 1. Prefight files above exist.
 2. `.cursor/context/config.json` has `externalPlanReview.enabled: true` (see `config.example.json`), **or** use `--force` for a one-shot arm without persisting opt-in. Missing file = disabled unless `--force`.
 3. Prefer `externalPlanReview.mode: "autonomous"` for background/inspectable auto-launch. Missing `mode` keeps paste-compatible / legacy behavior.
-4. Claude Code CLI (`claude`) on PATH for autonomous / interactive / headless launch; if missing, soft tip + exit 0 (Field Report stays owed). `--paste-only` still prints the command without requiring `claude` yet.
+4. A usable reviewer: `backend: "auto"` (default for new example/docs) uses Claude when present, else Cursor Agent. Pinned `backend: "claude"` still tips + no-op when Claude is missing. `--paste-only` still prints the command without requiring a binary yet. Same-family implementer and reviewer is an honest skip (including Auto/Auto). Claude review uses `reviewerModel` (default `haiku`); `advisorModel` (default `opus`) runs only on escalate.
 
 ## Manual arm
 
@@ -51,11 +51,12 @@ If any are missing: stop. Do **not** claim a review ran. Tell the user to run `a
 
 ### What "operator-visible" means (smoke notes)
 
-- **Autonomous success:** chat arm **must** use `--force --autonomous --wait-monitor`. The launcher prefers a background/inspectable PTY (no OS Terminal focus by default; `--focus-terminal` / `AGENT_KIT_AUDIT_FOCUS_TERMINAL=1` restores activate), then polls until a **fresh** monitor exists (`mtime >= arm epoch` or the HTML comment sentinel `<!-- audits-wait-fresh: created|updated -->` written into the monitor). Exit `0` = fresh ready; `3` = timeout; `4` = soft-fail while waiting. Spawn-only exit 0 without wait is **not** review done. **Chat continuation:** AwaitShell until `0|3|4`; on `0` run `/plan-review-triage` Ask in the same session. Do **not** stop at Final HANDOFF "after monitor lands" or require typing `done`. ADR: `decisions/2026-07-27_audits-wait-freshness-enforce.md`.
+- **Autonomous success:** chat arm **must** use `--force --autonomous --wait-monitor`. The launcher prefers a background/inspectable PTY (no OS Terminal focus by default; `--focus-terminal` / `AGENT_KIT_AUDIT_FOCUS_TERMINAL=1` restores activate), then polls until a **fresh** monitor exists (`mtime >= arm epoch` or the HTML comment sentinel `<!-- audits-wait-fresh: created|updated -->` written into the monitor). Chat AwaitShell uses `waitSliceSeconds` (default 90); total budget is `waitTimeoutSeconds` (default 900) persisted in `.cursor/context/audit-wait/<slug>.json`. A new session resumes remaining budget and does not restart 900s. Early-ready still exits `0` at first freshness. Exit `0` = fresh ready; `3` = timeout (slice or total; not review done); `4` = soft-fail while waiting. Spawn-only exit 0 without wait is **not** review done. **Chat continuation:** AwaitShell until `0|3|4`; on `0` run `/plan-review-triage` Ask in the same session. Do **not** stop at Final HANDOFF "after monitor lands" or require typing `done`. ADRs: `decisions/2026-07-27_audits-wait-freshness-enforce.md`, `decisions/2026-08-13_audits-atomic-wait-reviewer-fallback.md`.
 - **Autonomous soft-fail:** missing `claude` → tip + exit `4` when `--wait-monitor` was requested (Field Report owed). Background spawn unavailable → falls back to `--paste-only` UX with an honest "NOT running yet" banner. A **silent PTY** (spawn succeeded, no scrollback within the progress-gate grace window) is reported as a failed launch: the launcher disposes the session it just spawned, prints the paste fallback, and soft-fails instead of burning the wait budget. A **session-cap refusal** (detached `agent-kit-audit-*` sessions at the cap) never spawns at all. Soft-fail does **not** invent a monitor or run triage as if review completed.
 - **Exit 3 is timeout-only:** it means the freshness gate was not satisfied inside the budget, never that the review finished. A monitor that appears later, including one written by a different or later arm, does **not** convert a `3` into success. Leave the target Field Report **owed** and re-arm. ADR: `decisions/2026-07-30_audits-pty-progress-gate-zombie-policy.md`.
 - **Paste-only:** clipboard + printed interactive one-liner; review starts only after the operator pastes into their Cursor Terminal. After paste (Claude running), the session still waits for the monitor file then continues into triage Ask when possible.
-- **`--dry-run`:** resolves mode/plan and prints `background-cmd` / `paste-cmd` / `focus-terminal` without spawning Claude (useful for smoke).
+- **`--dry-run`:** resolves mode/plan and prints `background-cmd` / `paste-cmd` / `focus-terminal` / `reviewer-backend` / `same-model-refuse` without spawning a reviewer (useful for smoke).
+- **Reviewer cascade:** `--backend auto|claude|cursor` (or config). Claude spawn passes `--model` from `reviewerModel`. Cursor fallback cannot honor a Claude-family name; Auto/Auto is refused. Findings-only until `/plan-review-triage`. Never `/git-prod`.
 
 ### A. Script (preferred)
 

.cursor/commands/run-plan-all.md

@@ -184,13 +184,13 @@ Read `externalPlanReview` before the queue confirm Ask and at each advance:
 | Config | Behavior |
 |--------|----------|
 | Audits **pre-flight** (`preflight`: `off` \| `warn` \| `block`) | Before the confirm Ask and before each mid-queue advance: same owed/untriaged check as `/run-plan`. `block` arms or stops; never steals `/git-prod`. |
-| `midBatchAudits: true` and audits enabled | After each plan Task returns `outcome: completed`, the **orchestrator** arms **one** full audit for that plan with `--force --autonomous --wait-monitor` (or one `--batch` + wait_all when batching is intentional) **before** advancing the cursor. No paste Ask between plans. Soft-fail → Field Report owed; still advance. AwaitShell until exit `0|3|4`; wait success requires a **fresh** monitor after arm start. Do **not** fan out N background sessions without wait. Do **not** insert a mid-queue triage Ask (operator non-stop preserved; record ready path for queue-end). Mid-batch stays findings-only: **never** auto-Write residuals or rewrite the Run queue between plans. |
+| `midBatchAudits: true` and audits enabled | After each plan Task returns `outcome: completed`, the **orchestrator** arms **one** full audit for that plan with `--force --autonomous --wait-monitor` (or one `--batch` + wait_all when batching is intentional) **before** advancing the cursor. No paste Ask between plans. Soft-fail → Field Report owed; still advance. AwaitShell until exit `0|3|4` (chat slice ~90s; remaining budget in `.cursor/context/audit-wait/<slug>.json`). Wait success requires a **fresh** monitor after arm start. Reviewer cascade: `backend: "auto"` uses Claude (Haiku) when usable, else Cursor Agent. Same-model implementer/reviewer is an honest skip. Do **not** fan out N background sessions without wait. Do **not** insert a mid-queue triage Ask (operator non-stop preserved; record ready path for queue-end). Mid-batch stays findings-only: **never** auto-Write residuals or rewrite the Run queue between plans. |
 | `midBatchAudits` false/missing | **Non-stop** mid-queue: do **not** pause for audit Ask/paste between plans. Mid-queue completed plans stay Field Report **owed** until reviewed. |
 | Queue exhausted | Final HANDOFF; cadence `batch-complete`; then queue-end audit arm covering remaining owed/unreviewed targets (enabled → `--force --autonomous --wait-monitor` or paste per `mode`; else `offerOnExhausted` Ask). Prefer one launcher `--batch` + wait_all when multiple basenames. After wait exit `0`: run `/plan-review-triage` Ask with an **explicit path list** of fresh monitors (batch uniform Ask when outcomes match; sequential fallback when mixed; durable heading per file). **Batch exhaust without conveyor:** when remaining monitors are process-only / depth-capped, prefer uniform **Ack and stop** or **Fix nits only**; do not spawn unbounded `close-*` backlog from Write residuals (ADR `decisions/2026-08-11_plan-audit-residuals-termination.md`). Then suggest `/git-prod` if staging is ahead of `main` (separate HITL). |
 
 Never steal `/git-prod` confirmation. Chat never runs silent headless `--force` / `claude -p` in the agent shell. Spawn-only exit 0 without `--wait-monitor` is **not** review done. Never stop at Final HANDOFF "when monitors exist, run triage" after arming: wait (freshness) then continue (mid-batch waits for file only; queue-end waits then triage Ask with explicit paths). ADR: `2026-07-27_audits-autonomous-plan-review-contract.md` (supersedes queue-end-only); wait freshness: `2026-07-27_audits-wait-freshness-enforce.md`.
 
-**Exit 3 stays timeout-only across the queue.** A mid-queue or queue-end arm that returns `3` reviewed nothing: leave that plan Field Report **owed**, keep its path out of the queue-end triage list, and never narrate it as reviewed. Monitors that show up later, including monitors written by a different arm or a later queue position, do **not** retroactively upgrade an earlier `3`. Exit `4` covers the launcher soft-fails: missing `claude`, background spawn unavailable, a **silent PTY** early abort (spawn succeeded but produced no scrollback in the grace window), and a **session-cap refusal** (detached `agent-kit-audit-*` pile at the cap, so nothing spawned). Advance the queue on soft-fail, but record the target as owed, never as reviewed. ADR: `2026-07-30_audits-pty-progress-gate-zombie-policy.md`.
+**Exit 3 stays timeout-only across the queue.** A mid-queue or queue-end arm that returns `3` reviewed nothing: leave that plan Field Report **owed**, keep its path out of the queue-end triage list, and never narrate it as reviewed. Monitors that show up later, including monitors written by a different arm or a later queue position, do **not** retroactively upgrade an earlier `3`. Exit `4` covers the launcher soft-fails: no usable reviewer (`backend: "auto"` tried Claude then Cursor; pinned `claude` still tips when Claude is missing), same-model refuse, background spawn unavailable, a **silent PTY** early abort (spawn succeeded but produced no scrollback in the grace window), and a **session-cap refusal** (detached `agent-kit-audit-*` pile at the cap, so nothing spawned). Advance the queue on soft-fail, but record the target as owed, never as reviewed. ADR: `2026-07-30_audits-pty-progress-gate-zombie-policy.md`.
 
 ### External plan review (legacy heading)
 

.cursor/commands/run-plan-loop.md

@@ -11,4 +11,4 @@ When invoked, follow [`run-plan.md`](run-plan.md) forcing the **in-session loop*
 
 > "Heads up: `/run-plan-loop` is now `/run-plan`. Running with the in-session strategy."
 
-The full tick contract (plan status per tick, HANDOFF, automatic `/git-staging` on diff, **never** `/git-prod`, stop conditions) lives in `run-plan.md`, including the headless runner (`scripts/plan-loop.sh`) section.
+The full tick contract (plan status per tick, HANDOFF, automatic `/git-staging` on diff including HANDOFF/memory kit buckets, **never** skip those as trivial, **never** `/git-prod`, stop conditions) lives in `run-plan.md`, including the headless runner (`scripts/plan-loop.sh`) section.