メニュー
Answers questions about LangGraph and agentic AI using the langgraph-docs-mcp server. Use when the user asks about LangGraph, agentic AI, agent graphs, tool use, or graph-based workflows.
Codex または Claude でインストール この Prompt をコピーして Codex、Claude、または他のアシスタントに貼り付けると、Skill ページを確認してインストールできます。
SOC 職業分類に基づく
Adds a new TextMate-based language to the Monaco editor and Shiki highlighter in apps/ui, including precompiling grammars from repos/shiki, wiring codeLanguages and the contribution registry, and keeping artefacts traceable. Use when adding Monaco or Shiki support for a new file extension, creating a custom language pack, or copying the SysML v2 integration pattern.
Add new tools to the AI chat system. Use when adding a chat tool, creating tool schemas, wiring backend tool handlers, or building tool UI components.
Audit Cursor IDE context-bar consumption (rules, tools, skills, MCP, AGENTS.md) and recommend cuts. Use when the user complains rules/tools/AGENTS.md are taking too much context, opens the context bar and asks why a layer is large, asks to reclaim context budget, or asks to audit Cursor configuration for token bloat.
Programmatic and empirical UI audit of a deployed Tau environment (taucad.dev staging or tau.new production) covering Core Web Vitals, accessibility (WCAG 2.2 AA), security headers, SEO crawlability, bundle/network profile, and console errors. Use when the user asks for a UI audit, performance audit, accessibility audit, Lighthouse audit, axe-core audit, taucad.dev / tau.new audit, or wants to systematically validate a deployed environment against the vision/UI/accessibility/UX policies.
Create or update policy documents in docs/policy/. Use when writing a new policy, updating an existing policy, reviewing policy structure, or when the user mentions policy docs, coding standards, or architectural decisions that should be documented as policy.
Create or update research documents in docs/research/. Use when investigating a bug root cause, auditing code or configuration, comparing libraries or approaches, designing architecture, evaluating migration paths, or when the user mentions research, investigation, audit, analysis, or deep dive.
| name | langgraph |
| description | Answers questions about LangGraph and agentic AI using the langgraph-docs-mcp server. Use when the user asks about LangGraph, agentic AI, agent graphs, tool use, or graph-based workflows. |
for ANY question about LangGraph, use the langgraph-docs-mcp server to help answer --
