vibe-extras

Context-gathering agent for finding files to read (not analysis). Maps codebase structure; main agent reads files and reasons. Returns overview + prioritized file list with line ranges. Thoroughness: quick for "where is X?" lookups | medium for specific bugs/features | thorough for multi-area features | very-thorough for architecture/security audits. Auto-selects if not specified: single entity → quick; single subsystem → medium; multi-area → thorough; "comprehensive"/"all"/"architecture" → very-thorough. <example> prompt: "quick - where is the main entry point?" returns: Key files only, no research file </example> <example> prompt: "find files for payment timeout bug" returns: Payment architecture overview + prioritized files </example> <example> prompt: "very thorough exploration of authentication" returns: Dense auth flow overview covering all aspects </example>

Use this agent when you need to find and remove AI-generated code slop, including useless comments, verbose patterns, unnecessary abstractions, and filler phrases. This agent is ideal for cleaning up code after AI-assisted coding sessions, during code review to catch AI patterns, or when refactoring to improve code clarity. Examples: <example> Context: User just finished implementing a feature with AI assistance and wants to clean it up. user: "I just finished implementing the user authentication module, can you clean up any AI slop?" assistant: "I'll use the slop-cleaner agent to analyze and clean up AI-generated patterns in your authentication module." <Task tool call to slop-cleaner agent> </example> <example> Context: User wants to review recent changes for AI patterns before committing. user: "Check my recent changes for AI slop" assistant: "I'll launch the slop-cleaner agent to analyze the git diff between your branch and main to identify and remove any AI-generated patterns." <Task tool call to slop-cleaner agent> </example> <example> Context: User notices verbose comments in a specific file. user: "The utils.ts file has too many obvious comments, clean it up" assistant: "I'll use the slop-cleaner agent to remove useless comments and verbose patterns from utils.ts." <Task tool call to slop-cleaner agent with 'utils.ts' argument> </example>

Use this agent when you need to research external topics via web search - technology comparisons, best practices, industry trends, library evaluations, API documentation, or any question requiring current information from the web. The agent uses structured hypothesis tracking to systematically gather and synthesize web-based evidence. <example> Context: User needs to evaluate technology options. user: "What are the best options for real-time sync between mobile and backend in 2025?" assistant: "I'll use the web-researcher agent to systematically research and compare current real-time sync approaches." </example> <example> Context: User needs current best practices. user: "What's the recommended way to handle authentication in Next.js 15?" assistant: "Let me launch the web-researcher agent to gather current best practices and official recommendations." </example> <example> Context: User needs market/industry research. user: "What are the leading alternatives to Stripe for payment processing?" assistant: "I'll use the web-researcher agent to research and compare payment processing options." </example>

Build a personalized PR review skill by mining GitHub review history. Extracts review patterns, communication style, priorities, and blind spots — generates a calibrated review-as-me skill that posts inline comments on PRs. Use for onboarding, cloning review style, or creating a reviewer persona.

Find and remove AI-generated slop (useless comments, verbose patterns, unnecessary abstractions).

Find all files relevant to a query with orthogonal exploration for comprehensive coverage. Returns topic-specific overview + file list with line ranges. Uses parallel agents for thorough+ levels to ensure nothing is missed.

Autonomously bring a PR to mergeable state. Resolves merge conflicts, fixes CI failures, addresses review comments, syncs PR description, and requests re-review. Designed for unattended cron execution via loop skill. Use when: "prep pr for merge", "make pr mergeable", "fix pr blockers", "autonomous pr maintenance".

Update main/master from origin, rebase current branch on it, resolve conflicts, and push.

Deep web research with parallel investigators, multi-wave exploration, and structured synthesis. Spawns multiple web-researcher agents to explore different facets of a topic simultaneously, launches additional waves when gaps are identified, then synthesizes findings. Use when asked to research, investigate, compare options, find best practices, or gather comprehensive information from the web.\n\nThoroughness: quick for factual lookups | medium for focused topics | thorough for comparisons/evaluations (waves continue while critical gaps remain) | very-thorough for comprehensive research (waves continue until satisficed). Auto-selects if not specified.

Triage and fix PR review comments. Reads comments, filters false positives (outdated, misread, wrong context, style preference), validates relevance against current diff, and fixes valid ones with confirmation. Use when: "fix pr comments", "address review feedback", "handle pr feedback".

Rewrite branch into clean, narrative-quality commits. Creates backup, reimplements on fresh branch, verifies byte-identical, then replaces original branch history.

Create or update CLAUDE.md with best practices - brevity, universal applicability, progressive disclosure

Babysit a PR until merge-ready, then ask before merging. Runs pr-merge-prep on a 10-minute loop, monitors readiness (CI, reviews, conflicts), auto-stops when ready, and requires explicit user confirmation to merge. Never merges autonomously. Use when: "babysit this pr", "watch pr until ready", "keep fixing pr until mergeable".