菜单
Research an external website, API, tool, or service and produce an MMPS-specific feasibility + integration plan
用 Codex 或 Claude 帮你安装 复制这段 Prompt，粘贴到 Codex、Claude 或其他助手里，让它检查 Skill 页面并帮你完成安装。
基于 SOC 职业分类
| name | integration-research |
| description | Research an external website, API, tool, or service and produce an MMPS-specific feasibility + integration plan |
Investigate a third-party website, API, SDK, MCP, or platform the user is considering, and report whether/how it can be integrated into MMPS — with a concrete, convention-aware plan. This is research + planning only; it does NOT write feature code (hand off to /scaffold-service or /scaffold-ai-tool once the user approves).
Use this when the user says "look at this site/API and tell me what I'd need to do", "can I use X for this project?", "explore X, I want to add it", "investigate moving to X".
Pin down what the user actually wants from the integration (e.g. "get tweets of specific users", "generate images", "host the app cheaper"). Note any hard constraints they mention (free tier, no paid API, must run on a schedule, etc.).
Map the integration onto the repo's actual structure:
src/services/{name}/ (use /scaffold-service). Note the env var(s) needed.src/shared/ai/tools/ (use /scaffold-ai-tool).*-scheduler.service.ts cron job in the relevant feature.connectGithubMcp is wired in chatbot.init.ts).Output (in chat, or as an artifact in the session files/ dir if long) with these sections:
src/services/..., src/shared/ai/tools/..., schedulers, env vars, and the closest existing pattern to copy.Verifies factual claims in documents using web search and official sources, then proposes corrections with user confirmation. Use when the user asks to fact-check, verify information, validate claims, check accuracy, or update outdated information in documents. Supports AI model specs, technical documentation, statistics, and general factual statements.
Remove signs of AI-generated writing from text. Use when editing or reviewing text to make it sound more natural and human-written. Based on Wikipedia's comprehensive "Signs of AI writing" guide. Detects and fixes patterns including: inflated symbolism, promotional language, superficial -ing analyses, vague attributions, em dash overuse, rule of three, AI vocabulary words, passive voice, negative parallelisms, and filler phrases.
Plan new MMPS features by exploring the codebase and producing a structured implementation plan that follows project conventions
Complete browser automation with Playwright. Auto-detects dev servers, writes clean test scripts to /tmp. Test pages, fill forms, take screenshots, check responsive design, validate UX, test login flows, check links, automate any browser task. Use when user wants to test websites, automate browser interactions, validate web functionality, or perform any browser-based testing.
Generates optimized prompts for AI tools. Activates only when the user explicitly asks to write, fix, improve, or adapt a prompt for a specific AI tool (LLM, Cursor, Midjourney, image AI, video AI, coding agents, etc.). Does not activate for general conversation, coding tasks, document writing, or other non-prompt-engineering work.
Check changed code against MMPS project conventions and report violations with fix suggestions