brainstormer

Audit a React/TSX repository for React anti-patterns by loading rule cards from the brainstormer card library, scanning via each card's declared detect strategy, and filing one grouped GitHub issue per `(skill, rule_id)` via the `gh` CLI. Phase 3 adds the full re-run lifecycle: `findIssueByLabel` lookup, in-place body rewrite bounded by `react-audit:managed:start/end` sentinels (human comments survive), close-with-dated-resolution on emptied findings, regression-with-backlink on resurfaced findings, label-collision protocol under concurrent runs, never-reopen invariant. Phase 2c grouped body (15 MVP cards, per-occurrence severity, `<details>` collapsibles) and Phase 2b smart scan (`SMART_SCAN_THRESHOLD=50`) unchanged. Manual invocation only — no hooks. Triggers on `/react-audit`, "audit this repo for react anti-patterns", "run a react audit". Skip design questions, scaffolding, audits of non-React frameworks (Vue, Svelte, Solid).

Discovery interview and requirements gathering session. Pressure-test an idea, architecture, or design decision through structured client interviews. Use when the user wants to be challenged on their thinking, explore trade-offs, or vet a plan before committing.

Create a Product Requirements Document through structured interview. Use when the user wants to write a PRD, define requirements, or plan a new feature.

Turn a PRD into a multi-phase implementation plan using tracer-bullet vertical slices, saved as a local Markdown file in ./plans/. Use when user wants to break down a PRD, create an implementation plan, plan phases from a PRD, or mentions "tracer bullets".

Convert a PRD into independently-grabbable GitHub issues using vertical slices. Use when the user wants to create issues from a PRD, break down a PRD into tasks, or create GitHub issues from requirements.

Test-Driven Development workflow using vertical slices. Use when implementing features with TDD, writing tests before code, or doing red-green-refactor cycles.

Tell the agent to zoom out and give broader context or a higher-level perspective. Use when you're unfamiliar with a section of code or need to understand how it fits into the bigger picture.

Design and audit CLIs that AI agents (Claude Code, Codex, automated pipelines) consume. Applies ten agent-native principles in two tiers — Tier 1 table stakes (non-interactive default, structured output, errors that enumerate, safe retries, bounded responses) and Tier 2 compounding (cross-CLI vocabulary, three-layer introspection, async-aware execution, persistent profiles, two-way I/O) — with a Blocker / Friction / Optimization severity rubric across six command classes (read, mutate, stream, bootstrap, async, config). Use when building, reviewing, or specifying a CLI for agent invocation, mentions argparse, click, cobra, clap, commander, yargs, oclif, thor, agent-friendly or agent-native CLI design, --json output, exit codes, idempotent commands, --wait flags, agent-context introspection, persistent profiles, --deliver routing, --force bypass, vocabulary consistency, or repos with cli.py, main.go, src/cli.rs, bin/<tool>. Skip general shell scripting and human-only TUIs.

Capture the entire current chat as a structured handoff document the user can paste into Codex, Cursor, or another IDE-based AI coding agent. Triggers on /handoff, /export-context, /dump, "package this for codex", "send this to cursor", "make a handoff doc", "i wanna take this to another model", or similar phrasing indicating intent to continue work elsewhere. Produces lossless distillation — every meaningful detail survives, with discarded approaches explicitly labeled by reason so the next model does not re-suggest rejected paths.

Capture the entire current chat as a structured handoff document the user can paste into Codex, Cursor, or another IDE-based AI coding agent. Triggers on /handoff, /export-context, /dump, "package this for codex", "send this to cursor", "make a handoff doc", "i wanna take this to another model", or similar phrasing indicating intent to continue work elsewhere. Produces lossless distillation — every meaningful detail survives, with discarded approaches explicitly labeled by reason so the next model does not re-suggest rejected paths.

Orchestrated planning workflow — guides through discovery, PRD creation, vertical-slice phasing, and GitHub issue generation. Use when starting a new project or feature from scratch.

Audit and improve CLAUDE.md files by wrapping conditionally-relevant sections in <important if> tags, removing linter territory, and ensuring project accuracy. Use when user wants to optimize their CLAUDE.md for better agent performance.

Run a question through 5 AI advisors who analyze independently, peer-review anonymously, then synthesize a verdict. Invoke for strategic decisions with genuine uncertainty where multiple perspectives add value.

Lint, type-check, commit, and push in one flow. Use when user says /ship or asks to commit+push.

Enforces MVP development principles — speed over perfection, utility first, simplicity. Use proactively when planning features, architecting solutions, reviewing code, or evaluating technical decisions to prevent over-engineering.