rarestg
13 verified skills3 total stars
agents-md
Create, rewrite, or maintain a repository's AGENTS.md as a short dispatcher for coding agents. Use when asked to write an AGENTS.md from scratch, update stale agent instructions, shrink a bloated AGENTS.md, add repo guidance for AI/coding agents, or document where agents should find source-of-truth docs, invariants, validation commands, and workflow rules. Triggers on: "create AGENTS.md", "write an agents file", "update AGENTS.md", "maintain AGENTS.md", "fix our agent instructions", "make this repo easier for Codex/Claude/agents", "rewrite this bloated AGENTS.md".
development1
github-review-workflow
Export a PR's clean inline review comments, CodeRabbit outside-diff comments, and CodeRabbit nitpicks into local files, then triage review feedback through a stack-aware orchestrator workflow with durable reply drafts for follow-up PRs. Use when given a GitHub pull request URL and asked to work through review comments without relying on noisy raw API blobs.
development1
stacked-diffs
Break large code changes into small, stacked pull requests using vanilla git and the gh CLI. Auto-trigger when implementing a feature or change that spans multiple logical steps, touches several files, or will exceed ~200 changed lines. Also trigger on "stack PRs", "break this into smaller PRs", "stacked diffs", or "create a PR stack". Do NOT trigger for single-file fixes, small bug fixes, or changes under ~200 lines that are a single logical unit.
tools1
install-skills
Install, discover, remove, and update agent skills using the npx skills CLI. Use when asked to install a skill, add a skill from a repo, find or search for skills, list installed skills, remove or uninstall a skill, update skills, or check for updates. Triggers on: "install X skill", "add the Y skill", "find skills for Z", "what skills are available", "remove skill", "update my skills", "check for skill updates", "search for a skill that does X".
tools
skill-review
Review an agent skill for quality, completeness, and correctness. Evaluates purpose, approach, implementation, frontmatter, progressive disclosure, and whether scripts or references are needed. Use after creating or modifying a skill.
testing
execplan
Write a self-contained execution plan (ExecPlan) for complex, multi-step tasks. Use when asked to plan a large feature, significant refactor, multi-hour implementation, or when the user says "write a plan", "create an execplan", "plan this out", "design before building", or needs a structured approach before starting complex work.
development
task-os
Operating protocol that uses Taskwarrior as the continuity layer across conversations. Grounds every interaction in real task state before acting. Activate on "morning" / "briefing" for daily overview, on "status" / "what's next" for current state, when capturing tasks ("remind me to", "add task", "I need to"), when recovering from stalls, or when executing structured work across projects, repos, and life domains.
data-ai
coding-agent
Orchestrate Codex CLI and Claude Code as background assistants via tmux. Think of them as your team — delegate freely. Use when you want to: get a second opinion on your work; delegate research or context-heavy tasks; fan out work in parallel (map-reduce across agents); run the same instruction across many items; fix issues in parallel with worktrees; or build something end-to-end (research → plan → phased implementation → review). Triggers on: "run codex", "use claude on", "spawn an agent", "second opinion", "delegate", "fan out", "review my work", "build with agents".
tools
graphify-architecture-review
Run reusable Graphify-led architecture analysis for codebases using semantic graphs, optional subagent extraction, graph synthesis, source-search validation, graph-shape review, and follow-up refactor planning. Use when asked to analyze repo architecture, god nodes, surprising edges, topology, module boundaries, or graph-derived cleanup/refactor opportunities.
development
delegated-delivery
Run a meaningful coding ticket through a delegated delivery workflow: tighten the ticket, assign one ticket owner, delegate implementation, get an independent review, scan for high-value simplification, validate the result, and return a compact outcome packet. Use when the user wants structured agent execution with clear scope, ownership, and review rather than a single-pass implementation. Triggers on: "delegate this ticket", "use a sub-PM", "run this through worker and reviewer", "own this ticket end to end", "send this for independent review", or "close this ticket out with review". Skip trivial fixes and tasks that are still too vague to delegate.
testing
merge-stack
Merge a linear stack of GitHub PRs into main one by one. Use when the user has multiple PRs that form a chain and wants to merge them sequentially. Triggers on phrases like "merge the stack", "merge my PRs", or "merge stacked PRs".
data-ai
cf-browser
Browse and scrape websites using Cloudflare's Browser Rendering REST API. Use when the agent needs to fetch rendered web content, extract structured data from pages, take screenshots, or scrape specific elements via CSS selectors. Triggers on tasks like "scrape this site", "get listings from this page", "extract data from this URL", "take a screenshot of this page", "browse this website", or any task requiring headless browser access to read, crawl, or extract information from live web pages. Also use when WebFetch is insufficient (JS-heavy sites, SPAs, pages requiring cookies, or when structured extraction is needed).
development
code-quality
Set up formatting, linting, import sorting, type checking, and pre-commit hooks when scaffolding or starting a new project. Use this skill whenever creating a new project, initializing a repo, scaffolding an app, or when the user asks to add linting/formatting to an existing project. Triggers on: "new project", "scaffold", "init", "set up linting", "add formatter", "add pre-commit hooks", "configure biome", "configure ruff". The goal is to establish code quality tooling from day one so issues are caught incrementally, not in a painful bulk-fix later.
tools