Skip to content

ndisisnd/msg / .claude/skills/plan-pm

plan-pm

Verified

Principal PM skill — the autonomous PRD writer. Consumes a graded, fleshed-out row from the INTAKE.md backlog (idea, goal, type, grade) and drafts the full PRD solo — feature/acceptance table, edge cases, error handling — saved to features/planned/prd-[n]-[feature-slug]/. The requirements interview lives in /intake now, not here. Treats the row as settled and picks its own few follow-up questions (refs/principles.md); pauses ONLY for that one batch and for breaking/critical touches. --update edits an existing PRD's content and syncs its intake row; --migrate converts old PRDs to the current template. --review [on|off] runs plan-review at the end (this run, or saved per project). Refuses requests that would skip the PRD stage entirely.

FileKindSize
SKILL.mdskill md12.3 KB
refs/error-checklist.mdmarkdown2.8 KB
refs/principles.mdmarkdown2.4 KB
refs/protocol-migrate.mdmarkdown8.1 KB
refs/protocol-pm.mdmarkdown17.4 KB
refs/protocol-sub.mdmarkdown6.7 KB
refs/protocol-update.mdmarkdown6.9 KB
refs/template-prd.mdmarkdown11.8 KB

refs/error-checklist.md

View on GitHub
---
name: Error Checklist
description: The standard list of failure types plan-pm checks every §3 feature against when it drafts §4 Error and edge cases — Tier 1 always, Tier 2 only when the trigger fits — plus the two drafting rules that keep the table short.
type: reference
---

# Error Checklist

Check **every §3 feature** against this list when you draft §4. The list is a check, not a quota: it makes sure no common failure is forgotten, and most types will not apply to a given feature.

The **Type** name is the label you write in the §4 ID cell (`E1 · Network`).

## Tier 1 — check every feature

| Type | Ask of the feature |
|---|---|
| Invalid input | Can the user enter or submit something empty, too long, or wrongly formatted? |
| Network | Can the network or the server fail while this feature runs? |
| Slow response | Does the user wait on something that can be slow? What do they see meanwhile? |
| Double submit | Can the user tap twice, retry, or resend, and create a duplicate? |
| Interrupted | Can the action stop halfway (app closed, connection lost) and leave a half-done state? |
| Empty and limits | What does the user see the first time, with no data, or at a maximum? |
| Permission | Can a permission or a sign-in be missing, denied, or expired? |
| User mistake | Can the user undo it? Is any action irreversible, and is it confirmed first? |

## Tier 2 — check only when the trigger fits

| Type | Check it when the feature… |
|---|---|
| External service | uses a third-party service |
| Conflict and sync | works across devices, is shared between users, or works offline |
| Missing item | opens an item from a link or a notification (the item can be gone) |
| Time and date | depends on days, deadlines, streaks, or schedules |
| Rate limit | lets the user send, post, or request repeatedly |
| Data limits | uploads files or handles large data |
| Abuse and safety | shows user content or lets users contact each other |
| Privacy | could show one user's data to another |
| Money | charges, refunds, or pays out |
| Version mismatch | changes stored data or an API that old app versions still use |
| Device limits | uses the camera, location, storage, or a new OS feature |
| Language and accessibility | adds new text or layout — only when the project supports more than one language or has an accessibility rule |

## Two drafting rules

1. **Write a row only when the error is likely and user-visible.** Skip a type that cannot happen in this feature. Skip it silently — never write an `N/A` row.
2. **Merge rows that share one user-visible behaviour.** One row may name several triggers (for example "network timeout or server error on save"). Give it the type that fits best.

Expect about 2–5 rows per feature. Every §4 row is read by pre-merge's PRD-consistency check and by the manual test plan, so extra rows cost run time downstream.
reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis.Stars: 469kUpdated Claude CodeCodexVerifiedInstall
diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness.Stars: 321kUpdated Claude CodeCodexVerifiedInstall
dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependenciesStars: 321kUpdated Claude CodeCodexVerifiedInstall
executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is availableStars: 321kUpdated Claude CodeCodexVerifiedInstall