AGENT SKILL

project-to-prd

Use when an existing repository needs current-state product documentation reverse-extracted into a PRD suite from repository evidence.

License 许可
MIT
最近更新
2026-08-07
Stars
★ 0
Forks
0

01 前言 Front Matter

name
project-to-prd
metadata
{ "author": "github-spec-kit", "source": "templates/passive-skills/project-to-prd/SKILL.md" }
description
Use when an existing repository needs current-state product documentation reverse-extracted into a PRD suite from repository evidence.
compatibility
Requires spec-kit project structure with .specify/ directory

03 SKILL.md 正文

Project To PRD

Use this passive skill to recognize existing-project PRD extraction work and route it to sp-prd-scan, then sp-prd-build.

Use sp-specify when the user wants to align a new or changed feature intent before planning. Use sp-prd-scan -> sp-prd-build when the user wants a professional current-state PRD suite reconstructed from the repository that already exists. sp-prd is deprecated and remains compatibility-only for older surfaces that have not been updated yet.

The canonical PRD lane is a heavy reconstruction workflow: sp-prd-scan runs with execution_model: subagent-mandatory, critical claims must reach L4 Reconstruction-Ready, and the scan package includes contract artifacts such as config-contracts.json before sp-prd-build compiles exports. sp-prd-build is build-only: it must compile from the scan package instead of starting a second repository scan, and it must block completion when critical evidence is still missing.

Required Behavior

  • Route existing-project reverse PRD requests to sp-prd-scan, then sp-prd-build, before repository inspection.
  • Treat sp-prd-scan -> sp-prd-build as the depth-aware current-state extraction workflow, not a flat repo summary pass.
  • Treat the canonical lane as heavy reconstruction work with subagent-mandatory scan execution, not leader-only summarization.
  • Treat sp-prd-build as a build-only compilation step. It must not reread the repository, and it must refuse completion when critical evidence gaps remain.
  • Treat sp-prd as a deprecated compatibility-only entrypoint that must route into the canonical sp-prd-scan -> sp-prd-build flow.
  • Ground the PRD in current implementation reality: code, docs, tests, routes, UI surfaces, service/API surfaces, configuration, data models, domain terms, PROJECT-HANDBOOK.md, and project cognition evidence when present.
  • Require capability triage before claiming the PRD suite is complete.
  • Use targeted evidence harvest for critical and high capabilities.
  • Keep critical capabilities visible until they are depth-aware and L4 Reconstruction-Ready rather than merely surface-covered.
  • Preserve the distinction between Evidence, Inference, and Unknown.
  • Treat .specify/prd-runs/<run-id>/workflow-state.md as the resumable state source for the PRD run.
  • Keep the default terminal state as completed PRD exports under .specify/prd-runs/<run-id>/.
  • Do not automatically hand off to sp-plan, sp-tasks, or implementation planning.

Routing Signals

Use sp-prd-scan, then sp-prd-build for requests like:

  • "write a PRD for this existing app"
  • "extract product requirements from this repo"
  • "document what this product currently does"
  • "turn the current implementation into a product requirements document"
  • "generate UI/service PRD docs from the codebase"

Route to sp-specify instead when the user is describing a new product idea or future change that is not grounded in an existing repository surface.

Output Expectations

The active workflow should produce a current-state PRD suite, typically:

  • .specify/prd-runs/<run-id>/workflow-state.md
  • .specify/prd-runs/<run-id>/prd-scan.md
  • .specify/prd-runs/<run-id>/coverage-ledger.md
  • .specify/prd-runs/<run-id>/coverage-ledger.json
  • .specify/prd-runs/<run-id>/capability-ledger.json
  • .specify/prd-runs/<run-id>/artifact-contracts.json
  • .specify/prd-runs/<run-id>/config-contracts.json
  • .specify/prd-runs/<run-id>/reconstruction-checklist.json
  • .specify/prd-runs/<run-id>/evidence/
  • .specify/prd-runs/<run-id>/master/master-pack.md
  • .specify/prd-runs/<run-id>/exports/prd.md

Mode-specific exports may include UI, service, flow, data, rule, or internal brief views when the repository evidence supports them. Completion requires capability triage, targeted evidence harvest, and depth-aware handling for product-defining capabilities.

04 同仓库更多 Skills

chenziyang110
chenziyang110/launchdeck

launchdeck-agent

Codex host adapter contract fixture for Launchdeck.

★ 0
chenziyang110
chenziyang110/launchdeck

launchdeck-agent

Claude host adapter contract fixture for Launchdeck.

★ 0
chenziyang110
chenziyang110/launchdeck

launchdeck-agent

Use this skill aggressively for local project lifecycle operations even when the user does not mention Launchdeck: start this project, run the dev server, dev, build, test, restart service, stop service, show logs, inspect ports, port is occupied, clean cache; and when the user explicitly asks to configure this local project for Launchdeck or create or generate a project-adapted `.launchdeck.yml`, including 为这个项目生成 Launchdeck 配置. Do not use for ordinary code edits, docs-only work, general explanations, API design discussion, or production deployment unless the user also asks to operate or configure a local project or service.

★ 0
chenziyang110
chenziyang110/launchdeck

zig-testing

Use ONLY when user explicitly requests Zig test code, built-in test runner configuration, test failure diagnosis, or coverage for Zig projects. Triggers: "write test zig", "zig test", "testing.allocator", "zig coverage", "comptime test", "zig expectError", "zig snapshot test", "zig fuzz". Do NOT activate for: general Zig coding without test context, build.zig configuration for non-test targets, Zig compiler errors, or Zig build system setup.

★ 0
chenziyang110
chenziyang110/launchdeck

webapp-testing

Toolkit for interacting with and testing local web applications using Playwright. Supports verifying frontend functionality, debugging UI behavior, capturing browser screenshots, and viewing browser logs.

★ 0
chenziyang110
chenziyang110/launchdeck

verification-before-completion

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

★ 0