@hegemonart/get-design-done 1.28.0 → 1.28.5

package/skills/turn-closeout/SKILL.md

@@ -9,107 +9,87 @@ tools: Read, Bash
 
 ## Role
 
-You are a deterministic **closeout** skill. You close the per-turn telemetry gap on runtimes that don't expose a Stop event (codex, gemini, and 11 others). You are a code-level mirror of `hooks/gdd-turn-closeout.js` (D-10): same conditions, same idempotence, same emitted event shape. The only difference: the JS hook emits the nudge as `additionalContext` via the harness; this skill prints the nudge directly to the user.
+You are a deterministic **closeout** skill. You close the per-turn telemetry gap on runtimes that don't expose a Stop event (codex, gemini, and 11 others). You are a code-level mirror of `hooks/gdd-turn-closeout.js` (D-10): same conditions, same idempotence, same emitted event shape. The only difference: the JS hook emits the nudge as `additionalContext` via the harness; this skill prints the nudge directly to the user. See `./reference/milestone-completeness-rubric.md` §"Task level" for the broader closeout discipline (what "turn complete" means within a stage).
 
-**When to invoke:** orchestrator skills (`/gdd:next`, `/gdd:design`, `/gdd:verify`) tail-call this skill as their final step before returning, so the user sees a closing nudge that matches what Claude Code users see via the Stop hook. Adoption is incremental — each orchestrator can wire the tail-call independently; the skill exists as a stable, callable surface today.
+**When to invoke:** orchestrator skills (`/gdd:next`, `/gdd:design`, `/gdd:verify`) tail-call this skill as their final step before returning. Adoption is incremental — each orchestrator can wire the tail-call independently; the skill exists as a stable, callable surface today.
 
 ## Invocation Contract
 
-- **Input**: none. Operates on `.design/STATE.md` and `.design/telemetry/events.jsonl` in the current working directory.
+- **Input**: none. Operates on `.design/STATE.md` and `.design/telemetry/events.jsonl` in cwd.
 - **Output**: at most one printed line — the nudge — or silent return.
-- **Latency budget**: ≤10ms typical (matches D-10). Read **only** STATE.md and the tail of events.jsonl; never load the full event stream.
-- **Idempotence**: if the most recent event line is already a `turn_end` for the current `(stage, task_progress)` tuple, skip the append but still print the nudge.
-- **Non-blocking**: any I/O failure → silent return. This skill must never gate the user.
+- **Latency budget**: ≤10ms typical. Read **only** STATE.md and the tail of events.jsonl.
+- **Idempotence**: if the most recent event is already a `turn_end` for the current `(stage, task_progress)` tuple, skip append but still print nudge.
+- **Non-blocking**: any I/O failure → silent return. This skill never gates the user.
 
 ## Algorithm
 
-Execute these steps **in order** and stop at the first early-return.
+Execute in order; stop at the first early-return.
 
-### Step 1 — Try to read STATE.md
+### Step 1 — Read STATE.md
 
-Read `.design/STATE.md`. If the file is missing or unreadable: **return silently** (no print, no append). Mirrors the JS hook's "missing STATE.md" branch.
+Read `.design/STATE.md`. Missing or unreadable → **return silently** (no print, no append).
 
-### Step 2 — Parse the `<position>` block
+### Step 2 — Parse `<position>` block
 
-Lightweight-parse only the `<position>…</position>` block (the rest of STATE.md is irrelevant here). Extract `stage`, `status`, `task_progress`. A regex pass (`/<position>([\s\S]*?)<\/position>/` then per-line `key: value`) is sufficient — do **not** invoke the full STATE parser (cost overhead).
+Lightweight-parse only `<position>…</position>` (regex `/<position>([\s\S]*?)<\/position>/` then per-line `key: value`). Extract `stage`, `status`, `task_progress`. Do not invoke the full STATE parser.
 
-If `status != "in_progress"`: **return silently**. The pipeline is either initialized, completed, or blocked — no turn-end gap to close.
+If `status != "in_progress"` → **return silently**. Pipeline is initialized / completed / blocked — no turn-end gap to close.
 
 ### Step 3 — Tail the last event line
 
-Read **only the last 8 KiB** of `.design/telemetry/events.jsonl` (a single event line is ≪64 KiB). Treat all of these as "stale by definition":
-- The file is missing.
-- The file is empty.
-- The last line fails to parse as JSON.
-- The last line's `timestamp` is missing or unparseable.
+Read **only the last 8 KiB** of `.design/telemetry/events.jsonl`. Treat as "stale by definition":
 
-Otherwise compute `now - last_event.timestamp`. If the gap is **<60 seconds**, the user is actively mid-turn — **return silently** (the next real event will close the gap naturally).
+- File missing.
+- File empty.
+- Last line fails JSON parse.
+- Last line's `timestamp` is missing or unparseable.
 
-A reasonable Bash one-liner for the tail when running this skill in a runtime that lacks a Read-tail primitive: `tail -n 1 .design/telemetry/events.jsonl 2>/dev/null`.
+Otherwise compute `now - last_event.timestamp`. Gap < 60 seconds → user is actively mid-turn → **return silently** (next real event closes the gap naturally). Bash one-liner for the tail: `tail -n 1 .design/telemetry/events.jsonl 2>/dev/null`.
 
 ### Step 4 — Idempotence check, then append
 
-If the last event is already shaped `{type: "turn_end", stage: <same>, payload: {task_progress: <same>}}` for the **exact** `(stage, task_progress)` tuple from Step 2: **skip the append** but proceed to Step 5.
+If last event is already `{type:"turn_end", stage:<same>, payload:{task_progress:<same>}}` for the exact `(stage, task_progress)` from Step 2: **skip append** but proceed to Step 5.
 
-Otherwise append a single JSONL line to `.design/telemetry/events.jsonl`:
+Otherwise append one JSONL line to `.design/telemetry/events.jsonl`:
 
 ```json
 {"type":"turn_end","timestamp":"<ISO 8601 now>","sessionId":"<session-id-or-'turn-closeout'>","stage":"<stage>","payload":{"task_progress":"<N/M>"},"_meta":{"source":"gdd-turn-closeout-skill"}}
 ```
 
-Create `.design/telemetry/` if missing. The append must be a single `appendFile`-equivalent call (the writer assumes append-atomicity per Plan 20-06).
+Create `.design/telemetry/` if missing. Append is a single `appendFile`-equivalent call (writer assumes append-atomicity per Plan 20-06).
 
 ### Step 5 — Print the nudge
 
 Match `task_progress` against `^(\d+)/(\d+)$`:
 
-- **Numerator equals denominator and denominator > 0** (e.g. `5/5`, stage-complete):
+- **Numerator equals denominator and denominator > 0** (e.g. `5/5`): `Stage <stage> complete — run /gdd:next or /gdd:reflect`.
+- **Otherwise** (mid-task, e.g. `3/7`, `0/0`, malformed): `Stage <stage> paused mid-task — resume with /gdd:resume`.
 
-  > Stage `<stage>` complete — run `/gdd:next` or `/gdd:reflect`
-
-- **Otherwise** (mid-task, e.g. `3/7`, `0/0`, malformed):
-
-  > Stage `<stage>` paused mid-task — resume with `/gdd:resume`
-
-Print exactly one of these two lines. No additional commentary, no explanations of what the skill did — the nudge is the user-facing surface.
+One line exactly. No commentary — the nudge is the user-facing surface.
 
 ## Failure Modes
 
-Every step above has an explicit silent-return on failure. The skill must remain non-blocking under all conditions:
-
 | Condition | Behavior |
 |-----------|----------|
-| `.design/STATE.md` missing or unreadable | Silent return |
-| `<position>` block absent or malformed | Silent return |
+| STATE.md missing/unreadable | Silent return |
+| `<position>` absent or malformed | Silent return |
 | `status != "in_progress"` | Silent return |
-| `.design/telemetry/events.jsonl` missing | Treat as stale → fall through to append + nudge |
-| Last event line unparseable | Treat as stale → fall through |
-| Last event timestamp <60s old | Silent return |
-| Append fails (permission, disk full) | Print the nudge anyway; do not surface the I/O error |
-| Any uncaught throw at any step | Silent return |
+| `events.jsonl` missing | Treat as stale → fall through to append + nudge |
+| Last event unparseable | Treat as stale → fall through |
+| Last event < 60s old | Silent return |
+| Append fails (permission, disk full) | Print nudge anyway; do not surface I/O error |
+| Any uncaught throw | Silent return |
 
 ## Equivalence with the JS hook
 
-This skill and `hooks/gdd-turn-closeout.js` MUST stay code-level equivalent. Specifically:
-
-- Same four early-return branches (no STATE / not in_progress / fresh event / no-op).
-- Same staleness threshold: **60 seconds**.
-- Same idempotence guard: `(type=turn_end, stage, payload.task_progress)` triple.
-- Same emitted event shape (only `_meta.source` differs: `gdd-turn-closeout` vs `gdd-turn-closeout-skill`, so reflector telemetry can distinguish hook-driven vs skill-driven turn-ends).
-- Same nudge wording for both `N/N` and mid-task cases.
-
-If you change one, change the other in the same plan. Plan 25-09's `tests/turn-closeout-hook.test.cjs` covers the JS hook; the parallel coverage for this skill rides on Plan 25-09's Phase 25 baseline.
+This skill and `hooks/gdd-turn-closeout.js` MUST stay code-level equivalent: same four early-return branches, same 60-second staleness threshold, same idempotence guard (`type=turn_end, stage, payload.task_progress`), same event shape (only `_meta.source` differs: `gdd-turn-closeout` vs `gdd-turn-closeout-skill` so reflector telemetry can distinguish hook-driven vs skill-driven). Same nudge wording for both N/N and mid-task cases. Change one → change the other in the same plan. Plan 25-09's `tests/turn-closeout-hook.test.cjs` covers the JS hook.
 
 ## Non-Goals
 
-- **Not a state writer.** This skill never edits STATE.md. The events.jsonl append is the only side effect.
-- **Not a stage transition.** A `turn_end` event is a within-stage observation, not a state-machine move; downstream tools that gate on stage transitions ignore it.
-- **Not a Stop-event harness.** Cross-runtime Stop-event support at the harness level is explicit out-of-scope for Phase 25 (see CONTEXT.md OOS section).
+- **Not a state writer.** Never edits STATE.md. The events.jsonl append is the only side effect.
+- **Not a stage transition.** `turn_end` is within-stage, not a state-machine move; downstream tools ignore it for transition gating.
+- **Not a Stop-event harness.** Cross-runtime Stop-event support at the harness level is out-of-scope for Phase 25 (CONTEXT.md OOS).
 
 ## Integration Point
 
-The canonical tail-call sites (per D-11) are `/gdd:next`, `/gdd:design`, `/gdd:verify`. Each orchestrator's final step, immediately before returning to the user, should be:
-
-> Invoke skill `gdd-turn-closeout`.
-
-Tail-call wiring is intentionally not part of v1.25 (Plan 25-04 ships only the callable surface). Each orchestrator can adopt the wiring independently in a follow-up.
+Canonical tail-call sites per D-11: `/gdd:next`, `/gdd:design`, `/gdd:verify`. Each orchestrator's final step before returning to the user invokes `gdd-turn-closeout`. Tail-call wiring intentionally not part of v1.25 (Plan 25-04 ships only the callable surface).

package/skills/undo/SKILL.md

@@ -3,6 +3,7 @@ name: gdd-undo
 description: "Safe design change revert. Uses git log to find design commits, checks dependencies, reverts safely."
 argument-hint: "[<commit SHA>]"
 tools: Read, Write, Bash, AskUserQuestion
+disable-model-invocation: true
 ---
 
 # /gdd:undo

package/skills/update/SKILL.md

@@ -3,6 +3,7 @@ name: gdd-update
 description: "Update get-design-done to the latest release. Preserves .design/config.json and ./.claude/skills/."
 argument-hint: "[--dry-run] [--version <tag>]"
 tools: Read, Write, Bash
+disable-model-invocation: true
 ---
 
 # gdd-update