Pesquisa e Web

Use when the user wants to review their recent Claude Code sessions for patterns — analyzes the last N sessions (default 5) in the current project, dispatching parallel reviewers per session, then synthesizing cross-session findings

Use when creating or updating CLAUDE.md files for projects or subdirectories - covers top-level vs domain-level organization, capturing architectural intent and contracts, and mandatory freshness dates

Use when invalid data causes failures deep in execution - validates at every layer data passes through to make bugs structurally impossible rather than temporarily fixed

Use when writing instructions that guide Claude behavior - skills, CLAUDE.md files, agent prompts, system prompts. Covers token efficiency, compliance techniques, and discovery optimization.

Use when creating specialized subagents for Claude Code plugins or the Task tool - covers description writing for auto-delegation, tool selection, prompt structure, and testing agents

Use when creating or developing anything, before writing code or implementation plans - refines rough ideas into fully-formed designs through structured Socratic questioning, alternative exploration, and incremental validation

Use when the user wants to export a Claude Code session transcript as a readable Markdown file — converts the current session (or a specified transcript path) into GitHub-flavored Markdown with metadata header, collapsible tool results, and thinking blocks

Use to decide what kind of generic agent you should use

Use when starting feature work that needs isolation from current workspace or before executing implementation plans - creates isolated git worktrees with smart directory selection and safety verification

Use when planning features and need current API docs, library patterns, or external knowledge; when testing hypotheses about technology choices or claims; when verifying assumptions before design decisions - gathers well-sourced, current information from the internet to inform technical decisions

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

Use when implementing any feature or bugfix, before writing implementation code - write the test first, watch it fail, write minimal code to pass; ensures tests actually verify behavior by requiring failure first