菜单
Central utility skill for GSD operations. Provides config parsing, slug generation, timestamps, path operations, and orchestrates calls to other specialized skills. Acts as the unified entry point that the original gsd-tools.cjs provided via its lib/ modules (commands, config, core, init).
默认使用会先检查来源的 Prompt；你也可以切换为直接命令，或下载本地副本。
决定是否安装前，请先阅读 SKILL.md，以及 SkillsMP 当前展示的配套文件。
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
直接命令不会经过审查 Prompt；运行前请先检查来源。
npx skills add https://github.com/a5c-ai/babysitter --skill gsd-tools命令会保持在同一行。复制前请横向滚动并检查完整内容。
想先保存到本地？可下载 SkillsMP 当前能够提供的文件。
Reference for querying the Atlas knowledge graph through its MCP tools — the SECONDARY enrichment/comparison layer that adds best-practice context to systems you have ALREADY scanned from your real sources (`az`, repos, dirs). Use when you need to look up nodes, edges, kinds, clusters, stats, or wiki pages in Atlas to compare against your real inventory. (atlas graph, query atlas, atlas mcp, search the graph, graph neighbors, atlas record, atlas kinds, enrichment layer)
Atlas turns your STATED NEED into a real systems atlas by SCANNING your actual sources (Azure via `az`, git repos, local dirs) and process/data mining them, THEN enriching against the Atlas knowledge graph. Use this skill when asked to inventory/map your real systems, scan your cloud + repos + directories, mine the real processes or data they contain, or collect their real constraints/gotchas. (atlas, scan my systems, inventory our azure account, map my repos, real systems atlas, process mining, data mining, collect nuances, system discovery)
This skill should be used when the user asks to "find skills in the wild", "assimilate popular workflows", "discover SKILL.md files in repos", "research external skills", "find workflow patterns", "survey the skill landscape", "what skills exist out there", or wants to investigate public repositories for extractable processes, babysitter plugins, and reusable procedural insights. Searches GitHub for SKILL.md files, classifies repos by archetype, and maintains structured research under docs/reference-repos/.
基于 SOC 职业分类
正在显示 SKILL.md
| name | gsd-tools |
| description | Central utility skill for GSD operations. Provides config parsing, slug generation, timestamps, path operations, and orchestrates calls to other specialized skills. Acts as the unified entry point that the original gsd-tools.cjs provided via its lib/ modules (commands, config, core, init). |
| allowed-tools | Bash(*) Read Write Edit Glob Grep |
| metadata | {"author":"babysitter-sdk","version":"1.0.0","category":"gsd-core","backlog-id":"SK-GSD-001"} |
| graph | {"domains":["domain:software-engineering"],"skillAreas":["skill-area:agentic-loops","skill-area:orchestration-loop"],"workflows":["workflow:feature-development"],"topics":["topic:developer-experience"],"roles":["role:tech-lead","role:backend-engineer"]} |
.planning/config.json configuration.planning/ directory structure.planning/quick/.planning/ directory structureLoad and parse .planning/config.json with defaults:
{
"profile": "balanced",
"autoCommit": true,
"autoVerify": true,
"planCheckerEnabled": true,
"maxPlanRevisions": 2,
"waveParallelization": true,
"contextMonitor": true,
"contextWarningThreshold": 70,
"contextCriticalThreshold": 85,
"summaryVariant": "standard",
"tddEnabled": false,
"questioningDepth": "adaptive"
}
Read config:
cat .planning/config.json
If config does not exist, create with defaults. Merge user overrides with defaults (user values take precedence).
Generate kebab-case slugs from descriptions:
Input: "Add user authentication with OAuth2"
Output: "add-user-authentication-with-oauth2"
Input: "Fix bug #123 in payment processing"
Output: "fix-bug-123-in-payment-processing"
Rules:
-2, -3 if neededFormat timestamps for GSD artifacts:
Header: "2026-03-02T14:30:00Z"
Frontmatter: "2026-03-02"
Filename: "20260302-143000"
Display: "Mar 2, 2026 2:30 PM"
Resolve paths within .planning/ structure:
Phase directory: .planning/phase-{N}/
Plan file: .planning/phase-{N}/PLAN-{M}.md
Summary file: .planning/phase-{N}/SUMMARY.md
Quick task: .planning/quick/{NNN}-{slug}/
Debug session: .planning/debug/{slug}.md
Codebase docs: .planning/codebase/
Research docs: .planning/research/
Milestone archive: milestones/v{X}.{Y}/
Parse integer and decimal phase numbers:
"72" -> { major: 72, minor: null, display: "72" }
"72.1" -> { major: 72, minor: 1, display: "72.1" }
"72.2" -> { major: 72, minor: 2, display: "72.2" }
Decimal phases are inserted phases (e.g., urgent work between phase 72 and 73).
Sequential numbering for quick tasks:
.planning/quick/001-fix-login-bug/
.planning/quick/002-add-rate-limiting/
.planning/quick/003-update-readme/
Scan existing directories, find highest number, increment by 1. Zero-pad to 3 digits.
Ensure .planning/ directory exists with required structure:
.planning/
config.json
PROJECT.md (if new-project has run)
REQUIREMENTS.md (if new-project has run)
ROADMAP.md (if new-project has run)
STATE.md (always present after init)
quick/ (created on demand)
debug/ (created on demand)
codebase/ (created on demand)
research/ (created on demand)
Detect current project state for routing:
const states = {
NO_PROJECT: "No .planning/ directory or no PROJECT.md",
INITIALIZED: ".planning/ exists with PROJECT.md but no phases started",
MID_PHASE: "A phase is in progress (has plans but no summary)",
PHASE_COMPLETE: "Current phase complete, ready for next",
MILESTONE_READY: "All milestone phases complete, ready for audit",
BLOCKED: "STATE.md has active blockers"
};
Read to load .planning/config.jsonGlob to check for duplicates in target directoryBash to create directory structure with mkdir -pWrite to create config.json with defaults if missingRead + Glob to detect existing stateThis skill is used by all GSD processes as the first initialization step:
new-project.js - Initialize .planning/, create configplan-phase.js - Resolve phase paths, load config for planner settingsexecute-phase.js - Load config for executor settings, resolve task pathsquick.js - Generate quick task number and slug, resolve quick task pathdebug.js - Generate debug session slug, resolve debug pathprogress.js - Detect project state, load configcomplete-milestone.js - Resolve milestone archive pathadd-tests.js - Resolve phase paths for test generationresearch-phase.js - Resolve research output paths{
"operation": "init|config|slug|path|state",
"status": "success|error",
"result": {
"config": {},
"slug": "generated-slug",
"path": "/resolved/path",
"state": "NO_PROJECT|INITIALIZED|MID_PHASE|..."
},
"metadata": {
"planningDir": ".planning/",
"configPath": ".planning/config.json",
"timestamp": "2026-03-02T14:30:00Z"
}
}
| Setting | Default | Description |
|---|---|---|
profile | balanced | Model profile (quality/balanced/budget) |
autoCommit | true | Auto-commit after each task |
autoVerify | true | Auto-verify after phase execution |
planCheckerEnabled | true | Run plan checker before execution |
maxPlanRevisions | 2 | Max plan revision cycles |
waveParallelization | true | Enable wave-based parallel execution |
contextMonitor | true | Enable context window monitoring |
summaryVariant | standard | Summary template variant |
| Error | Cause | Resolution |
|---|---|---|
Config parse error | Malformed config.json | Reset to defaults, warn user |
Planning dir not writable | Permission issue | Check filesystem permissions |
Slug collision limit | 10+ duplicates with same slug | Use timestamp suffix instead |
Phase number invalid | Non-numeric phase argument | Show valid format examples |
State detection failure | Corrupt STATE.md | Suggest health command to repair |
.planning/ directory without explicit instruction