12 skills · sorted by relevance
Design or change REA investigation tools by defining responsibilities, inputs, results, and CLI or MCP interfaces. Produce a proposal that explains how Evidence records should be interpreted, compares the nearest existing alternative, and outlines verification.
Design tools that AI agents can choose and use correctly. Define clear descriptions, inputs, and outputs, manage large search results, and plan approval checks, isolated execution environments, and interfaces for collaboration between agents.
Plan and build an MCP server, a connection that lets Claude use external tools. Choose hosting, tool structure, authentication, and a framework, then scaffold a server or hand off to specialized packaging and interface skills.
Design tools that agents can select and call correctly. Clarify descriptions, input rules, response formats, and recovery messages, then review overlapping tools and naming conventions using examples of successful and failed calls.
Plan projects that use large language models, from checking whether the task suits a model to designing processing stages. Estimate costs, choose single or multiple agents, and organize prototypes, structured outputs, and iteration.
Designing side-by-side comparison tools (plan-compare, product-compare, alternative-compare) that help users decide rather than just listing features. Axis selection, default-comparison logic, recommendation discipline. Honest about feature-list-dump (every feature in a row, no decision support), hidden-recommendation (biased comparison pretending to be neutral), and honest-comparison-with-guidance (genuine comparison plus opinionated recommendation) patterns. Triggers on comparison tool, plan compare, product compare, alternative compare, vs page, decision support tool. Also triggers when conversion through comparison stages is poor, when users are abandoning at the comparison step, or when a comparison tool is being scoped for the first time.
The English summary is being prepared. Showing the original description.
Build MCP tools that are sticky for vibe coders and powerful for developersUse when "Designing new MCP tools, Improving tool UX or DX, Writing error messages, Planning tool naming, Discussing user onboarding, Making tools "sticky", Vibe coder experience, mcp, product, ux, dx, vibe-coding, onboarding, developer-experience, tool-design" mentioned.
The English summary is being prepared. Showing the original description.
Analyze an arbitrary codebase to identify functions, APIs, and data sources suitable for exposure as MCP tools, producing a tool specification document. Use when planning an MCP server for an existing project, auditing a codebase before wrapping it as an AI-accessible tool surface, comparing what a codebase can do versus what is already exposed via MCP, or generating a tool spec to hand off to scaffold-mcp-server.
The English summary is being prepared. Showing the original description.
The Agent Tool Contract — 5 principles for designing tools agents call reliably: predictable signature, rich errors, token-efficient output, idempotency, graceful degradation. Includes anti-pattern table with 8 common mistakes.
The English summary is being prepared. Showing the original description.
Design tools that agents can use effectively, including when to reduce tool complexity. Use when creating, optimizing, or reducing agent tool sets.
The English summary is being prepared. Showing the original description.
This skill should be used when the user asks to "design agent tools", "create tool descriptions", "reduce tool complexity", "implement MCP tools", or mentions tool consolidation, architectural reduction, tool naming conventions, or agent-tool interfaces.
The English summary is being prepared. Showing the original description.
Build tools that agents can use effectively, including architectural reduction patterns. Use when creating new tools for agent systems, debugging tool-related failures or misuse, or optimizing existing tool sets for better agent performance.
The English summary is being prepared. Showing the original description.