.claude/skills/brainstorm/SKILL.md
Structured pre-planning research. Explores codebase, asks clarifying questions, and produces a brainstorm output file for /plan-feature input.
npx skillsauth add lucidlabs-hq/agent-kit brainstormInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Run structured research BEFORE creating an implementation plan. Explore the codebase, ask the right questions, research options, and produce a brainstorm document that feeds into /plan-feature.
/plan-feature for complex featuresParse the argument as the brainstorm topic:
/brainstorm auth improvements
/brainstorm rate limiting strategy
/brainstorm convex schema redesign
Before asking questions, gather context:
# What exists related to the topic?
# Search codebase for relevant code
# Read related reference docs
# Check for prior brainstorms/reviews on this topic
Check prior work:
ls .agents/brainstorms/ | grep -i "{topic keywords}"
ls .agents/reviews/ | grep -i "{topic keywords}"
ls .agents/plans/ | grep -i "{topic keywords}"
Use AskUserQuestion to understand intent and constraints:
Question Categories:
| Category | Example Questions | |----------|------------------| | Scope | "Should this cover all API routes or just the new ones?" | | Constraints | "Any third-party libraries we should avoid?" | | Priority | "Is this for the pilot customer or all customers?" | | Timeline | "Is this blocking the next deploy?" | | Approach | "Do you prefer X approach or Y approach?" |
Rules:
Based on answers, research approaches:
.claude/reference/ docs say about this area?Write the output to:
.agents/brainstorms/YYYY-MM-DD-{topic-slug}.md
Document Template:
# Brainstorm: {Topic}
**Date:** YYYY-MM-DD
**Author:** Agent + {developer name}
**Status:** Complete -- Ready for /plan-feature
---
## Context
{What prompted this brainstorm. Link to Linear ticket if applicable.}
## Current State
{What exists today. Code references with file:line format.}
## Requirements (from Discussion)
{Answers from the clarifying questions.}
---
## Options Explored
### Option A: {Name}
**How:** {Description}
**Pros:** {List}
**Cons:** {List}
**Effort:** {S/M/L}
**Files touched:** {List}
### Option B: {Name}
**How:** {Description}
**Pros:** {List}
**Cons:** {List}
**Effort:** {S/M/L}
**Files touched:** {List}
---
## Recommendation
**Go with Option {X}** because {reason}.
{If no clear winner, explain trade-offs and let user decide.}
---
## Key Decisions Needed
- [ ] {Decision 1}
- [ ] {Decision 2}
---
## References
- {Link to relevant reference docs}
- {Link to relevant code}
- {External resources}
---
## Next Step
Run `/plan-feature {topic}` using this brainstorm as input.
After writing the document, show a compact summary:
BRAINSTORM COMPLETE
----------------------------------------------------------------------
File: .agents/brainstorms/YYYY-MM-DD-{topic}.md
Topic: {topic}
Options: {N} explored
Recommendation: Option {X} -- {one-line reason}
Decisions needed: {N}
Next: /plan-feature {topic}
----------------------------------------------------------------------
/brainstorm rate limiting
Explores rate limiting options, asks about scope, produces recommendation.
/brainstorm multi-tenant auth redesign
Deep research into current auth, explores BetterAuth patterns, researches alternatives.
/brainstorm {topic} -- Research and explore options
|
v
/plan-feature {topic} -- Design implementation plan (uses brainstorm as input)
|
v
/execute {plan} -- Implement the plan
|
v
/review -- Validate the implementation
The brainstorm document is automatically discovered by /plan-feature when the topic matches.
development
Deploy invoice-accounting-assistant to HQ server. Runs tests first (TDD), then builds and deploys. Use when ready to push changes to staging/production.
testing
Visual UI verification with agent-browser. Use after implementing UI components to take screenshots, verify interactions, and self-check your work. FASTER than E2E tests.
documentation
Update README with current project status and features. Use after completing features.
tools
--- name: time-report description: Cross-project time report. Aggregates all session data from ~/.claude-time/sessions/. Use to see how much time was spent across all projects. disable-model-invocation: true allowed-tools: Bash, Read argument-hint: [all | this-week | this-month | last-month | {project-name}] --- # Time Report: Cross-Project Session Overview ## Objective Read ALL session files from `~/.claude-time/sessions/*.json` and produce an aggregated time report. Supports filtering by pe