research-machine

Search multiple job sources for new listings, spawn parallel triage agents, and update tracking state. Use when running /job-discover or searching for new job opportunities.

Staff-engineer-level ADR helper — interactive multi-round exploration that helps you deeply understand architecture decisions through codebase research, ASCII diagrams, and Socratic dialogue. Use when you want to understand an ADR ("explain this ADR", "help me understand this decision", "what does this mean", "walk me through this"), review it ("is this ADR good enough", "poke holes", "review this ADR"), or when sharing ADR file paths (docs/decisions/, docs/adr/, numbered files like 0014-*.md). Every round produces diagrams and ends with user-directed next steps.

Curate weekly newsletter from checked stories in daily AI digests. Extracts [x] marked items, groups into narrative themes, generates editorial newsletter.

Comprehensive grounded research with rigorous source verification. Use when user needs thorough research with verbatim citations, fact-checking verification, or academic-level source grounding on complex topics.

Interactive KubeCon CFP submission writer. Guides through topic selection, title crafting, abstract writing, and benefits section using acceptance data from 1,100+ talks across 7 KubeCon events (2024-2025). Use when preparing a conference talk proposal for KubeCon/CloudNativeCon, writing a CFP, or asking about KubeCon submission strategy.

Critique a Claude implementation plan before approving execution. Spawns parallel persona subagents (Skeptic, Architect, Verifier) to evaluate the plan against the codebase and return an Approve/Reject/Refine verdict with concrete refinements. Use when a plan has been generated (Plan Mode output, ExitPlanMode, pasted plan text, or plan file path) and needs review before code is written. Triggers on "critique this plan", "review this plan", "is this plan good", "poke holes in this plan", "should I approve this plan", or sharing a plan for evaluation.

Quick web research with source citations. Use when user asks for brief research, fact-checking, quick lookups, or wants concise answers with sources on any topic.

Multi-source recipe research with metric conversion and verification

Generate new Claude Code skills from requirements through guided questions and pattern selection

Staff-engineer-level code review that goes beyond correctness to evaluate architectural alignment, system-level implications, failure modes, observability, security, and cross-team impact. Use when reviewing a PR (URL or diff), analyzing code changes for architectural fitness, or when the user asks for a thorough/staff-level/senior review of code changes. Triggers on "review this PR", "review these changes", "staff review", "thorough code review", sharing a GitHub PR URL for review, or asking about the architectural impact of changes.

Analyze repo for technical debt, research language-specific best practices, create umbrella tracker issue with one subissue per finding

Write tests that verify behavior (not implementation), use table-driven/parameterized patterns, and minimize mocking. Triggers when asked to write tests, add test coverage, or create test files. Also triggers when reviewing existing tests for quality.

Summarize a YouTube video from its transcript and save a concise note to the Obsidian vault. Use when user shares a YouTube URL and wants a summary, key takeaways, or notes from a video.

Analyze the current repo for technical debt, research version-specific best practices, and create an umbrella GitHub tracker issue with one subissue per finding. Use when the user asks to audit, find, or track technical debt in a code repository.

Generate project-specific CLAUDE.md with domain instructions, workflows, templates. Analyzes codebase, asks clarifying questions, produces tailored instructions.

Comprehensive Polish plot evaluation with full due diligence (~20-30 min). Use when user wants deep analysis of a property listing after initial triage, or needs complete risk assessment for a finalist plot.

Auto-review Go code for 100+ common mistakes when analyzing .go files, discussing Go patterns, or reviewing PRs with Go code. Checks error handling, concurrency, interfaces, performance, testing, and stdlib usage.