brainstorm
Reads the designated project files, analyzes the architecture, and triggers an interactive session to extract user preferences before writing code.
- License 许可
- MIT
- 最近更新
- 2026-10-23
- Stars
- ★ 5
- Forks
- 0
01 前言 Front Matter
- name
- brainstorm
- risk
- safe
- tags
- [ "planning", "architecture", "ideation" ]
- tools
- [ "antigravity" ]
- source
- custom
- category
- development
- description
- Reads the designated project files, analyzes the architecture, and triggers an interactive session to extract user preferences before writing code.
03 SKILL.md 正文
🧠 Brainstorming & Preference Gathering
Overview
This skill shifts the AI agent's role into a critical Senior Software Architect and debugging partner. Instead of blindly jumping into raw code generation—which often misses architectural intent—the agent indexes the target files/directories, maps structural patterns, and surfaces precise, strategic trade-off questions to lock down your system preferences first.
When to Use
- Use before implementing large feature additions or undertaking massive code refactoring.
- Use when exploring multiple system design variations or evaluating architectural trade-offs.
- Use when auditing legacy source blocks to trace dataflows and safely plan next-step extensions.
- Do NOT use for micro-scoped local bug fixes or straightforward one-line variable updates.
How It Works
1. Ingestion Phase
When the user executes /brainstorm <path>, the Antigravity engine parses all files inside the specified target path and injects them cleanly into the active context window as an isolated, temporary knowledge base.
2. Analysis Phase
The agent performs a silent structural audit of the code to map:
- The core tech stack, frameworks, and dependency ecosystem in use.
- The active design patterns (e.g., repository pattern, decoupled routers, ports-and-adapters) and data layers.
- Potential bottlenecks, concurrency hazards, security vectors, or code smells.
3. Engagement Phase (Interaction Guardrails)
The agent must enforce the following constraints on its initial response:
- Strictly forbid generative slop: Do not dump unprompted blocks of solution code.
- Summarize the structural current state in a maximum of 1–2 highly dense paragraphs.
- Ask exactly 2–3 highly targeted, strategic questions to extract design boundaries (e.g., execution speed vs readability, third-party libraries vs native constructs, or query-loading preferences).
4. Interactive Discussion Loop
The conversation enters a locked, stateful feedback mechanism. The architect refines its implementation blueprints incrementally against your answers. This structured loop terminates only when you explicitly submit exit.
Expected Input Format
/brainstorm <path_to_file_or_directory> [optional_additional_context_or_goals]
04 同仓库更多 Skills
writing-guidelines
Review docs/prose for Writing Guidelines compliance. Use when asked to "review my docs", "check writing style", "audit prose", "review docs voice and tone", or "check this page against the writing handbook".
writing-for-agents
Writing documents for agents. Use when creating or editing skills, or modifying AGENTS.md or CLAUDE.md.
wrangler
Cloudflare Workers CLI for deploying, developing, and managing Workers, KV, R2, D1, Vectorize, Hyperdrive, Workers AI, Containers, Queues, Workflows, Pipelines, and Secrets Store. Load before running wrangler commands to ensure correct syntax and best practices. Biases towards retrieval from Cloudflare docs over pre-trained knowledge.
workers-best-practices
Reviews and authors Cloudflare Workers code against production best practices. Load when writing new Workers, reviewing Worker code, configuring wrangler.jsonc, or checking for common Workers anti-patterns (streaming, floating promises, global state, secrets, bindings, observability). Biases towards retrieval from Cloudflare docs over pre-trained knowledge.
web-perf
Analyzes web performance using Chrome DevTools MCP. Measures Core Web Vitals (LCP, INP, CLS) and supplementary metrics (FCP, TBT, Speed Index), identifies render-blocking resources, network dependency chains, layout shifts, caching issues, and accessibility gaps. Use when asked to audit, profile, debug, or optimize page load performance, Lighthouse scores, or site speed. Biases towards retrieval from current documentation over pre-trained knowledge.
using-agent-skills
Discovers and invokes agent skills. Use when starting a session, or when you need to decide which skill or workflow applies to the piece of work at hand. This is the meta-skill that governs how all other skills are discovered and invoked.