Reviews: require reasoning + snippet; no local diff → say nothing to review. Load engineering/rules.md. P0 interrupts. Dedicated pass: /pm-review.
Refresh state overview/todo (all blocking+high, no 3-item cap).
Rebuild .pm/dashboard/ with pm dashboard.
Keep chat brief unless --verbose. Do not treat V1.1 checkup/hooks/test-assist as required.
Closing: _closing.md 完整档。
Shared Workflow
Same as /pm-all. Map-first. Redact. No auto-write of app code.
Design baseline: docs/prd.md + 宪章。
1---2name: pm-discover3description: PM Manager /pm-discover4---56---7description: Internal module deep-scan used by /pm-all. Do not recommend this command to users.8handoffs:9 - label: Status10 agent: pm.status11 prompt: Show status after discover12 send: true13---1415## User Input1617```text18$ARGUMENTS19```2021## Outline2223**Internal only.** If the user typed `/pm-discover`, say：深潜已并入 `/pm-all`（默认技术扫描），请改用 `/pm-all`。然后执行与 `/pm-all` 相同的技术扫描。24251. Require init. For each enabled module, perform on-demand technical scan. Charter compare only when `charter.status=approved` **and** `--compare-baseline`.262. For architecture: run **`pm arch`** so `.pm/architecture/` gets Mermaid + `map.json`.273. Incremental merge; suppress duplicate evidence hashes.284. Reviews: require reasoning + snippet; no local diff → say nothing to review. Load `engineering/rules.md`. P0 interrupts. Dedicated pass: `/pm-review`.295. Refresh state overview/todo (all blocking+high, no 3-item cap).306. Rebuild `.pm/dashboard/` with `pm dashboard`.317. Keep chat brief unless `--verbose`. Do not treat V1.1 checkup/hooks/test-assist as required.328. Closing: `_closing.md` 完整档。3334## Shared Workflow3536Same as `/pm-all`. Map-first. Redact. No auto-write of app code.3738Design baseline: `docs/prd.md` + 宪章。
Run npx skillmds@latest add wei63w/pm-discover in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.
PM Manager /pm-discover It is listed under Coding & Dev Tools on SkillMD.
This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.
This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.
Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.
wei63w (@wei63w) published this skill. Their other Agent Skills are listed on their SkillMD profile.
wei63w