repo-onboarding
repository discovery, entrypoint mapping, and fast onboarding for unfamiliar codebases. use when the repo is new, mixed, legacy, monorepo-shaped, poorly documented, or when the next safe step is blocked on understanding layout, ownership, and commands.
- License 许可
- GPL-3.0
- 最近更新
- 2026-05-19
- Stars
- ★ 6
- Forks
- 1
01 前言 Front Matter
- name
- repo-onboarding
- description
- repository discovery, entrypoint mapping, and fast onboarding for unfamiliar codebases. use when the repo is new, mixed, legacy, monorepo-shaped, poorly documented, or when the next safe step is blocked on understanding layout, ownership, and commands.
02 资源文件 7
- 参考文档references/command-discovery.md500 B
- 参考文档references/discovery-checklist.md565 B
- 参考文档references/examples.md504 B
- 参考文档references/monorepo-signals.md491 B
- 参考文档references/output-template.md455 B
- 参考文档references/ownership-and-health-signals.md614 B
- 参考文档references/repo-archetypes.md425 B
03 SKILL.md 正文
Repository Onboarding
Trigger Signals
- the repository is unfamiliar, mixed, legacy, or monorepo-shaped
- the next safe step is blocked on understanding layout, ownership, or workflow
- docs are sparse, stale, or scattered across many entrypoints
Inspect First
- repo root files such as
README.md, top-level manifests, CI config, container config, and deployment hints AGENTS.md,.agents/skills/,.specify/, or other repo-native guidance surfaces when present- top-level source, app, package, service, infra, and test directories
CODEOWNERS, contribution docs, issue or PR templates, and other health signals when present
Workflow
- identify the repo archetype and dominant implementation surfaces
- map build, test, run, deploy, and validation entrypoints
- infer ownership, boundaries, and high-risk hotspots from structure and metadata
- call out drift, missing validation, generated code, or conflicting conventions
- recommend the next best action and the best repo surfaces to load before changing code
Output Contract
- repository summary with dominant languages, runtimes, and repo shape
- key commands and entrypoints for build, test, run, and release
- hotspots, blockers, and missing information that affect safe changes
- recommended next action with the specific files or dirs to inspect next
Failure Modes
- repo layout is too large to inspect safely in one pass
- generated or vendored code hides the real ownership boundaries
- no reliable build or validation entrypoints can be found
Escalation
- escalate when ownership or architecture cannot be inferred safely
- escalate when there is no trustworthy validation path for a risky change
- escalate when docs and code disagree about the active workflow
References
skills/repo-onboarding/references/output-template.mdskills/repo-onboarding/references/discovery-checklist.mdskills/repo-onboarding/references/repo-archetypes.mdskills/repo-onboarding/references/ownership-and-health-signals.mdskills/repo-onboarding/references/command-discovery.mdskills/repo-onboarding/references/monorepo-signals.mdskills/repo-onboarding/references/examples.md
04 同仓库更多 Skills
typescript-engineering
typescript and javascript engineering guidance seeded from the harness forge knowledge base. use when the repo or task is dominated by .ts, .tsx, package.json, tsconfig, node.js services, react, next.js, frontend tooling, or multi-package workspaces where type boundaries matter.
token-budget-optimizer
compact context, reuse existing runtime artifacts, and choose the smallest authoritative surface before expanding prompt history. use when the task is growing large, when earlier repo intelligence already exists, or when repeated rescans would waste tokens without adding new evidence.
test-strategy-and-coverage
Define right-sized test layers and coverage priorities for a change.
speckit-taskstoissues
Convert existing tasks into actionable, dependency-ordered GitHub issues for the feature based on available design artifacts.
speckit-tasks
Generate an actionable, dependency-ordered tasks.md for the feature based on available design artifacts.
speckit-specify
Create or update the feature specification from a natural language feature description.