Skip to content

puncsky/system-design-and-architecture / .agents/skills/pm-brainstorm

pm-brainstorm

Verified

Propose milestones and tasks for the repository's public .pm board — decompose a topic through the editor's reader-impact lens, size and gate the work, and emit a text-only proposal for /pm to materialize. Use only when the user explicitly invokes /pm-brainstorm or asks to brainstorm, plan, or break down book roadmap work. Not for writing board files (that is pm) or ordinary chapter edits.

Task: Propose milestones and tasks

Usage: /pm-brainstorm <topic or goal to break down>

The divergent half of the pair: think a topic through, propose work, hand materialization to /pm. It writes nothing. Conventions (pillars B1/B2/B3, hierarchy, sizing, closing tasks, templates) live in ../pm/SKILL.md — read and apply them; do not restate or diverge.

Minimum proposal size: at least 5 distinct, actionable top-level candidates (milestones and/or inbox notes) that pass scope, anti-goal, dedup, sizing and quality rules. Don't pad or artificially split.

Steps

  1. Load canon and anti-goals. Read .agents/skills/pm/SKILL.md and .pm/DO_NOT_DO.md. Reject conflicting candidates explicitly.
  2. Sync if safe. git fetch origin, git status; fast-forward only when behind and clean. Never block on a dirty/diverged tree.
  3. Load context. Read the board (find .pm -name README.md, loose notes, done/ and blocked/) so proposals fit and don't duplicate shipped work. Ground truth for the book is the repo itself: README.md, SUMMARY.md, zh-CN/README.md, the relevant en/ chapters, drafts/, and awesome-system-design.md. Check that a candidate chapter doesn't already exist (ls en | grep, grep -il), and what the en/zh-CN parity gap actually is (diff <(ls en) <(ls zh-CN)).
  4. Apply the editor's lens. For each candidate answer: Pillar (B1/B2/B3, else out); Reader (interview candidate, practicing engineer, translator/contributor); Friction removed (what they can now learn/find/trust); Signal (site traffic, GitHub stars/issues, reader feedback, build passing). Prefer compounding assets (fixing a template, parity tooling, a TOC) over one-off prose; prefer fixing wrong/outdated content over adding more.
  5. Discuss & decompose. Pressure-test scope, surface dependencies/risks; break into tasks with estimates and depends_on.
  6. Size and gate by /pm's rules. Undersized → inbox note. Fails the gate → "not meaningful" with a better-scoped alternative.
  7. Emit text only. Full detail first: target queue and why, each milestone (task table, definition of done, source, pillar, outcome, why-now) or inbox note, in proposed priority order. Implementation tasks only — /pm appends closing tasks; flag whether you expect Reader surface to apply.
  8. End with exactly two blocks, nothing after:
    1. "Summary (priority order)" — N. <title> (wN, ~size) — one-line outcome.
    2. The exact /pm invocations (/pm new milestone w1 <title>, /pm add w1 <idea>, /pm promote w1/NNN).

Topic

$ARGUMENTS

slack-gif-creatoranthropics/skillsKnowledge and utilities for creating animated GIFs optimized for Slack. Provides constraints, validation tools, and animation concepts. Use when users request animated GIFs for Slack like "make me a GIF of X doing Y for Slack."Stars: 198kUpdated Claude CodeCodexVerifiedInstall
ponytailDietrichGebert/ponytailLazy senior dev mode for any coding task: the smallest change that fully solves it, and a reply a busy human understands in one read.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailHarvest every shortcut: comment into one debt ledger, so deferrals get tracked instead of forgotten. One-shot report.Stars: 194kUpdated Claude CodeCodexVerifiedInstall
ponytail-debtDietrichGebert/ponytailList every `shortcut:` comment (and older `ponytail:` ones) as a debt ledger. One-shot report, changes nothing. Use for "ponytail debt", "what did ponytail defer", "list the shortcuts", /ponytail-debt.Stars: 194kUpdated Claude CodeCodexVerifiedInstall