task
Single task creation shortcut. Use when /plan routes here (scope=TASK) or for quick task capture. For 2+ tasks, use /plan instead.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
메뉴
Single task creation shortcut. Use when /plan routes here (scope=TASK) or for quick task capture. For 2+ tasks, use /plan instead.
Codex 또는 Claude로 설치 이 Prompt를 복사해 Codex, Claude 또는 다른 어시스턴트에 붙여 넣으면 Skill 페이지를 검토하고 설치를 진행할 수 있습니다.
SOC 직업 분류 기준
Guide CLAUDE.md updates based on doc-guard suggestions. This skill should be used when updating documentation, fixing docs, addressing doc-guard suggestions, or updating CLAUDE.md. Activates when user requests "update documentation", "fix docs", "doc-guard suggestions", or "update CLAUDE.md".
Monitor blocked workers and help respond to them. Loop checks ft inbox, shows blocked workers, lets you respond naturally.
Adds tasks to FormalTask with mandatory discovery and validated criteria. Use when "add task", "create task", or "/task-add". For batch creation, use /epic-decompose instead.
Groups uncommitted changes into logical commits, walks through each grouping for approval, then optionally syncs with remote. Resolves merge conflicts interactively with context from recent remote and local history. Use when "commit my changes", "group commits", "smart commit", "commit and sync", or "organize my changes". Never resolves conflicts without user confirmation.
Unified critique for plans, specs, OR task artifacts. Auto-detects target. Spawns auditors. Commits tasks to DB when READY. Use "critique", "review", "is this ready".
Systematic debugging with graded exploration and agent escalation. MUST BE USED when encountering errors, test failures, or unexpected behavior.
| name | task |
| description | Single task creation shortcut. Use when /plan routes here (scope=TASK) or for quick task capture. For 2+ tasks, use /plan instead. |
| argument-hint | <title or description> |
| inherit | ["review"] |
| tools | ["auggie"] |
| required_todos | ["quick-discovery","validate-paths","validate-epic","specify","write-artifact"] |
mcp__auggie-mcp__codebase-retrieval(
information_request=f"Find code related to: {description}"
)
If 0 paths found → search broader. If 5+ files involved → "Scope too big. Run: /plan {description}" STOP
EXIT CRITERIA: 1-4 file paths with line numbers.
Spawn plan-explorer to verify:
Task(subagent_type="plan-explorer",
prompt=f"Validate these paths exist and contain expected code: {paths}")
ABSOLUTE PATHS ONLY in prompts. No ~/ or relative paths. Hooks enforce this.
EXIT CRITERIA: Paths confirmed valid. Invalid → re-run Phase 0.
ft epic list --names | grep -qx "$EPIC_NAME"
No epic specified? Default to inbox with confirmation:
AskUserQuestion(questions=[{
"question": "Add this task to inbox epic?",
"header": "Epic",
"options": [
{"label": "Yes, inbox", "description": "Quick capture, triage later"},
{"label": "No, different epic", "description": "I'll specify which epic"}
],
"multiSelect": False
}])
EXIT CRITERIA: Epic validated or defaulted to inbox.
Title: Action verb + specific scope.
REJECT these vague patterns:
properly, correctly, well, good, clean, improved, better,
appropriate, suitable, handles errors, is robust, is efficient,
works correctly, handles edge cases, is maintainable, follows best practices
Every criterion MUST be binary testable. Can't write a test? Rewrite it.
Reviews: Default code-quality. Add based on domain:
| Task Content | Add Review |
|---|---|
| Auth, credentials, user input | security |
| Database operations | sqlite |
| Loops, large data | perf |
| File paths | path-security |
| State transitions | state-machine |
Documentation required: true only for public API, CLI, user-facing changes.
EXIT CRITERIA: Title + 2-3 testable criteria + reviews + doc flag.
from pathlib import Path
from datetime import datetime, timezone
import re
date = datetime.now(timezone.utc).strftime("%Y-%m-%d")
slug = re.sub(r"[^a-z0-9]+", "-", title.lower()).strip("-")
artifact_dir = Path.home() / "projects" / "one-offs" / f"{date}-{slug}"
artifact_dir.mkdir(parents=True, exist_ok=True)
# Task: {title}
**Epic:** {epic_name}
**Required reviews:** {reviews}
**Documentation required:** {true|false}
## Goal
{One sentence}
## Acceptance Criteria
- [ ] {criterion_1}
- [ ] {criterion_2}
- [ ] No new classes unless unavoidable (plain functions preferred)
- [ ] Junior dev understands each function in 30 seconds
## Anti-patterns (DO NOT)
- No wrapper functions that just call another function
- No config options added "for flexibility"
- No "Manager" or "Handler" classes
## Files to Modify
- `{file}:{line}` - {change}
Output:
══════════════════════════════════════════════════════════════
✓ TASK: {title}
══════════════════════════════════════════════════════════════
Artifact: {artifact_dir}/task.md
Next: /critique {artifact_dir}
══════════════════════════════════════════════════════════════
EXIT CRITERIA: Artifact written. Run /critique to review and commit to DB.
- 5+ files = route to /plan, not task - MUST spawn plan-explorer to validate paths — no skipping - Reject any criterion containing vague patterns - Default to inbox epic with confirmation when no epic specified - No DB commit here — /critique reviews and commits when READY