deep-plan-ingest
把瞬态计划转成可供代理加载的验证过的项目知识，提升长期价值。
把深度计划文档提炼为 AGENTS/ARCHITECTURE/ROADMAP 等持久知识文件。
把一份深度计划（比如 docs/PLAN.md）里的结论提炼成团队能长期用的三份知识文件：AGENTS.md、ARCHITECTURE.md、ROADMAP.md。在计划完成后用，目的是把被验证过的事实留下、把过时或凭空的假设剔除，并以合并的方式更新而不是生硬覆盖。特别之处是只带入有证据支持的内容、会复查陈旧断言，确保后续的 agents 和工程文档能可靠加载这些知识。
▸ 展开 SKILL.md 英文原文
Distills a completed deep-plan document (e.g. docs/PLAN.md) into durable project knowledge files: AGENTS.md (agent operating rules and verified ground truth), docs/ARCHITECTURE.md (component map, data flow, framework mechanisms with citations), and ROADMAP.md (phased roadmap with live status). Carries only VERIFIED evidence forward, re-checks stale claims, and merges into existing files instead of overwriting. Use after a deep-plan finishes, when the user mentions deep-plan-ingest or ingest the plan, or asks to generate/update AGENTS.md, ARCHITECTURE.md, or ROADMAP.md from a plan.
帮我安装这个 skill：https://raw.githubusercontent.com/silkyland/deep-plan/main/skills/deep-plan-ingest/SKILL.mdcurl -fsSL "https://raw.githubusercontent.com/silkyland/deep-plan/main/skills/deep-plan-ingest/SKILL.md"# Deep Plan Ingest A plan document is a snapshot; it goes stale the day implementation starts. This skill moves the **verified knowledge** out of the plan and into the three files future agents actually load — so the research paid for once keeps paying. | Target | Purpose | Audience | |--------|---------|----------| | `AGENTS.md` | Operating rules + ground-truth facts for agents working in this repo | Every future agent session | | `docs/ARCHITECTURE.md` | Component map, data flow, framework mechanisms — with citations | Agents and humans exploring the system | | `ROADMAP.md` | The phased roadmap with live phase status | Anyone asking "what's next?" | ## The Prime Directive (inherited fro