oh-my-customcode 0.164.0 → 0.166.0

package/dist/cli/index.js

@@ -241,7 +241,7 @@ var init_package = __esm(() => {
     workspaces: [
       "packages/*"
     ],
-    version: "0.164.0",
+    version: "0.166.0",
     description: "Batteries-included agent harness for Claude Code",
     type: "module",
     bin: {

package/dist/index.js

@@ -2031,7 +2031,7 @@ var package_default = {
   workspaces: [
     "packages/*"
   ],
-  version: "0.164.0",
+  version: "0.166.0",
   description: "Batteries-included agent harness for Claude Code",
   type: "module",
   bin: {

package/package.json

@@ -3,7 +3,7 @@
   "workspaces": [
     "packages/*"
   ],
-  "version": "0.164.0",
+  "version": "0.166.0",
   "description": "Batteries-included agent harness for Claude Code",
   "type": "module",
   "bin": {

package/templates/.claude/agents/mgr-gitnerd.md

@@ -65,3 +65,24 @@ Types: feat, fix, docs, style, refactor, test, chore
 ## Push Rules (R016)
 
 All pushes require prior mgr-sauron:watch verification. If sauron was not run, REFUSE the push.
+
+## Milestone Query Robustness
+
+When verifying milestone state (e.g., confirming it is closed after a release), prefer **number-based direct query** over title-matching list lookup:
+
+```bash
+# Preferred: direct lookup by milestone number (deterministic)
+gh api repos/{owner}/{repo}/milestones/<number> --jq '.title, .state, .open_issues'
+
+# Fallback: title-matching list lookup (may fail transiently)
+gh api "repos/{owner}/{repo}/milestones?state=all&per_page=100" \
+  --jq '.[] | select(.title == "vX.Y.Z") | .title, .state, .open_issues'
+```
+
+**Rules:**
+- If title-matching list lookup returns no results (apparent "not found"), do NOT immediately report the milestone as absent.
+- Retry once (transient jq filter / pagination timing issues can cause false negatives).
+- If still not found after retry, fall back to number-based direct query before reporting "milestone does not exist."
+- False "milestone not found" reports can mislead the release milestone-close verification step.
+
+Origin: #1287 (v0.164.0 session retrospective — milestone v0.164.0 reported as absent but confirmed present via direct re-query).

package/templates/.claude/rules/MUST-completion-verification.md

@@ -196,6 +196,19 @@ triage-dispatch.yml 실패 원인을 파일 Read 전에 "triaged 라벨 부재 +
 
 Origin: #1266 ④.
 
+### Directory-Context Before Multi-Copy Unification/Deletion
+
+다중 사본(동일 파일이 N곳에 존재)을 통일하거나 삭제하기 전, 각 사본이 위치한 **디렉토리 전체 맥락**을 확인한다(`ls`로 형제 파일 파악). 사본 파일 하나만 보고 "orphan"·"stub"으로 특성화하면, 같은 디렉토리의 형제 파일(다른 역할을 가진)이 함께 덮이거나 맥락이 누락된다. Read-Before-Characterize를 파일 단위에서 디렉토리 단위로 확장한 규칙이다.
+
+| 금지 | 필수 |
+|------|------|
+| 사본 파일만 보고 "orphan/stub"으로 단정 후 통일/삭제 | 사본이 속한 디렉토리 전체(`ls`)를 확인 — 형제 파일 역할·연계 파악 후 처리 |
+
+#### Common Violation (#1290 찐빠 #2, cross-session)
+Session 108에서 `auto-dev.yaml` 4곳을 canonical 통일할 때, repo-root `./workflows/`에 `eraser.yaml`이 공존하는 디렉토리 맥락을 미확인하고 덮었다. Session 109에서 디렉토리 단위 Read-Before-Characterize로 보정(`eraser.yaml` 발견 → #1289 등록, destructive 삭제 회피). 결과는 무해했으나 맥락이 불완전했다.
+
+Origin: #1290 (session 109 retrospective).
+
 ### Degraded-Output Re-Verification Gate (529 / buffering)
 
 When tool outputs show degradation signs — 529 errors, duplicated or truncated output, or a Read returning empty on a file that is known non-empty — you MUST re-verify any fact via a deterministic second source BEFORE any destructive or permanent action (recovery-agent dispatch, issue edit, commit, file restore). Do NOT characterize state ("corruption", "오염", "loop") from a single degraded read.

package/templates/.claude/rules/MUST-orchestrator-coordination.md

@@ -317,6 +317,14 @@ Before delegating a task to a subagent, MUST verify the target agent's tool capa
 
 > **Path existence ≠ tool capability (#1269 ③)**: the pre-check above verifies the agent HAS Read/Write/Bash, but not that the target path actually exists. Delegating a read/write to a missing or renamed path causes the same round-trip waste the capability pre-check is meant to prevent. Verify path existence (Glob/ls) before delegating path-specific work.
 
+> **Multi-copy content consistency (#1287)**: 동일 파일이 다중 사본으로 존재하는 경우(예: auto-dev.yaml이 실행본 + templates 미러 + 레거시 사본 등 N곳), 위임 전 경로 존재뿐 아니라 **사본 간 내용 일관성(md5/diff)도 확인**해야 한다. 사본이 drift된 상태에서 "N곳 동일 변경 적용"으로 위임하면 에이전트가 작업 중에야 drift를 발견(round-trip)하거나, 일부 사본만 갱신되어 불일치가 심화된다.
+>
+> | Anti-pattern | Required |
+> |--------------|----------|
+> | `find`로 N곳 존재 확인 후 "N곳 동일 변경" 위임 | 위임 전 `md5`/`diff -q`로 N곳 내용 일치 확인; drift 시 canonical 기준 정렬을 위임 prompt에 명시 |
+>
+> Origin: #1287 (v0.164.0 세션 회고 찐빠 #1).
+
 ### Known Limitations (Active Cache)
 
 | Agent | Limitation | Workaround |

package/templates/.claude/rules/MUST-parallel-execution.md

@@ -32,6 +32,7 @@ Before writing/editing multiple files:
 3. Specialized agent available? → Use it (not general-purpose)
 4. Agent Teams available? → **Check R018 criteria before spawning 2+ agents**
 5. Running agent stalled (2x+ duration)? → Spawn independent follow-up tasks immediately
+6. Announced a parallel dispatch in prose? → ALL announced tool calls MUST be in the SAME message as the announcement (announce-execution consistency)
 
 ### Common Violations to Avoid
 
@@ -42,6 +43,9 @@ Before writing/editing multiple files:
 ❌ WRONG: Single agent receives massive multi-domain prompt (>5000 tokens, e.g., M2 plan with 12 tasks across 7 areas)
    → Latency timeout, user cancellation, context waste, no review loop
 ✓ CORRECT: Pre-decompose by domain, spawn parallel agents per area (R009) or use Agent Teams (R018)
+
+❌ WRONG: Announce "milestone 생성 + 구조 확인 병렬" but only dispatch one tool; the other runs next turn (announce-execution mismatch)
+✓ CORRECT: When announcing N parallel tools, include ALL N tool calls in the SAME message as the announcement
 ```
 
 > **Token threshold heuristic**: When a delegated agent prompt exceeds ~5000 tokens or spans 3+ unrelated domains, decompose by domain and spawn parallel agents. See R018 for Agent Teams criteria when review cycles are needed. Reference: #1085.
@@ -64,6 +68,8 @@ Before writing/editing multiple files:
 
 > **Agent Teams partial spawn** → See R018 (MUST-agent-teams.md) "Spawn Completeness Check".
 
+> **v2.1.161+**: Parallel tool calls in a single batch are now independent — a failed Bash command no longer cancels the other calls in the same batch; each tool returns its own result. This strengthens R009 batching: one failing call in a parallel dispatch no longer aborts its siblings, so independent work bundled in the same message completes regardless of a single failure. Lowers the safety cost of the announce-execution consistency self-check (#6).
+
 ## Execution Rules
 
 | Rule | Detail |

package/templates/.claude/rules/SHOULD-hud-statusline.md

@@ -19,6 +19,8 @@ Format: `─── [Spawn] {subagent_type}:{model} | {description} ───`
 
 > **v2.1.157+**: `tool_decision` telemetry events now include `tool_parameters` (bash commands, MCP/skill names) when `OTEL_LOG_TOOL_DETAILS=1`. Complements R012 observability — enables per-tool parameter tracking in monitoring dashboards. See `monitoring-setup` skill.
 
+> **v2.1.161+**: `OTEL_RESOURCE_ATTRIBUTES` values are now emitted as labels on metric datapoints — usage metrics can be sliced by custom dimensions (e.g., team, repo). Extends R012 observability from per-tool parameters (v2.1.157) to per-dimension metric slicing; configure via the `monitoring-setup` skill's `OTEL_RESOURCE_ATTRIBUTES` env. Separately, `claude agents` rows now show `done/total` progress before the detail when work is fanned out, and peek surfaces the longest-running item — complements the HUD parallel-spawn display and R009 `[N]` correlation.
+
 <!-- DETAIL: HUD Events full spec
 ### When to Display: Multi-step tasks, parallel execution, long-running operations. Skip for single brief operations.
 ### Parallel Display:

package/templates/.claude/skills/pipeline/SKILL.md

@@ -138,6 +138,22 @@ steps:
 }
 ```
 
+## Workflow File Locations
+
+The `auto-dev.yaml` (and other workflow YAML files) exist in **4 locations**. Only one is the runtime source — the others are mirrors for deployment and examples.
+
+| Path | Role | Used by |
+|------|------|---------|
+| `.claude/skills/pipeline/workflows/auto-dev.yaml` | **Runtime source (실행본)** — `/pipeline` reads from here via `Glob("workflows/*.yaml")` relative to the skill base dir | `/pipeline` skill at runtime |
+| `templates/.claude/skills/pipeline/workflows/auto-dev.yaml` | Init deployment mirror — copied to the runtime location when `omcustom init` runs | `omcustom init` |
+| `workflows/auto-dev.yaml` (repo root) | Legacy `/omcustom:workflow` directory remnant — NOT referenced by `/pipeline`. Also contains `eraser.yaml` and template examples. | Unused after `/pipeline` migration |
+| `templates/workflows/auto-dev.yaml` | Template example mirror — Glob'd by List Mode as "template examples" (`Glob("templates/workflows/*.yaml")`) | `/pipeline` List Mode display only |
+
+**Key rules:**
+- The runtime source is `.claude/skills/pipeline/workflows/` (skill base dir). Do NOT confuse with repo-root `workflows/`.
+- When modifying any workflow YAML, update **all applicable mirrors** to prevent drift. `verify-template-sync.sh` (#1286) detects drift automatically on CI.
+- Repo-root `workflows/` is a legacy `/omcustom:workflow` remnant — it also contains `eraser.yaml`. Do not delete without checking for other references.
+
 ## Error Handling
 
 - Pipeline not found → list available pipelines with suggestion

package/templates/manifest.json

@@ -1,5 +1,5 @@
 {
-  "version": "0.164.0",
+  "version": "0.166.0",
   "lastUpdated": "2026-05-20T00:00:00.000Z",
   "omcustomMinClaudeCode": "2.1.121",
   "omcustomMinClaudeCodeReason": "Sensitive-path direct Write/Edit on .claude/** under bypassPermissions (R010 deprecation, #1101)",