Coding & Dev Tools
Coding agent skills teach AI agents repeatable engineering workflows: reviewing pull requests, writing tests, refactoring safely, and enforcing house style. Install one SKILL.md and your agent applies the same checklist every time, whether you use Claude Code, Cursor, Codex, or another agent.
-
yeshwanthyk Skill SwarmRun a bounded parallel investigation or solution race and reconcile the results.
-
yeshwanthyk Skill RecallReconstruct project history and current state to answer a history question or resume work.
-
yeshwanthyk Skill UnslopClean AI-sounding prose while preserving meaning, facts, warnings, and the intended voice. Use for reports, logs, docs, skill text, review output, or any prose that feels padded or generic.
-
yeshwanthyk Bundle ReflectOn request, extract reusable lessons from a session and propose or apply scoped instruction changes.
-
yeshwanthyk Bundle ShapingDefine an unsettled problem and compare solution shapes with the user.
-
yeshwanthyk Bundle Show MeExplain the current topic visually; create a visual change outline whenever creating or updating a pull request description.
-
yeshwanthyk Bundle ArchitectChoose an architecture when a consequential design boundary is unsettled.
-
yeshwanthyk Skill Automate MeCreate or update a personal mode from explicit preferences and repeated working conventions.
-
yeshwanthyk Bundle InterrogateUse independent reviewers to challenge a design or change and return an evidence-checked judgment.
-
yeshwanthyk Bundle Yesh RouterOn explicit request, route work to the collection's methods and apply relevant engineering principles. Use when the user asks for yesh-router.
-
yeshwanthyk Skill Blast RadiusInvestigate what a change could break beyond its diff and prove the important safety claims.
-
yeshwanthyk Bundle Fuzz TestingBuild and run bounded coverage-guided fuzz targets for parsers, codecs, protocols, and untrusted-input boundaries; minimize and replay failures.
-
yeshwanthyk Bundle Property TestingBuild and run property-based tests for contracts spanning many inputs or operation sequences, with shrinking and replayable failures.
-
yeshwanthyk Bundle Structure ReviewReview code structure for concrete ownership, contract, and maintainability problems.
-
yeshwanthyk Bundle Show Me Your WorkKeep an evidence-linked decision trail for work that needs later review.
-
yeshwanthyk Bundle Create Verification SkillCreate and prove a project-local skill for driving a real application surface.
-
natejswenson Bundle AppletvInteract with and control the Apple TVs on the local network from the Mac you are on — scan for them, pair with one, and drive it: play, pause, skip, turn on or off, open an app or a deep link, type into the on-screen keyboard, navigate the menu, set the volume, and say what is playing — reading the device's state back after every command. Use when the user says "appletv", "control my apple tv", "connect to my apple tv", "find my apple tvs", "pair with the apple tv", "pause the apple tv", "turn off the tv", "what's playing on the apple tv", "open netflix on the living room tv", "skip the intro", "type stranger things into the tv", or wants to drive an Apple TV without picking up the remote. Built on pyatv over the Companion and AirPlay protocols.
-
natejswenson Bundle ReleaseCut a release for one named skill, package or service and prove the tag exists. Use when the user says "release devlog", "release ghostwriter", "cut a release for press", "ship a new version of X", "what is unreleased", "tag the current main", or asks whether something has a release pending. Reads the commits on main since the last tag, proposes a semver bump and a CHANGELOG entry, waits for approval, then lands the bump and reports the tag URL — never claiming a release happened until the tag is read back from the remote.
-
natejswenson Bundle IssueflowTake one GitHub issue through an independently reviewed plan, implementation, draft pull request, and converging review loop. Use when the user says "work an issue", "fix issue 42", "take this issue to a PR", or asks to list open issues. Runs autonomously by default; use --review-plan only when a human plan gate is explicitly requested.
-
natejswenson Bundle SkillhelpAnswer help questions about the skills in this repo. Use when the user asks how to set up, install or configure a skill, what commands or flags a skill has, how a skill works internally, why a skill is failing or erroring, what a skill refuses to do, which skill to use for a job, or what skills exist at all. Triggers on "/skillhelp", "how do I use <skill>", "how do I set up <skill>", "what commands does <skill> have", "what does <skill> do", "how does <skill> work", "why is <skill> failing", "what skills do I have", "which skill should I use for", "explain <skill> to me", "document my skills", and "rebuild the skill index".
-
natejswenson Bundle PluginsyncRefresh the locally-installed Claude Code plugin marketplaces so newly released skill versions are actually on disk. Use when the user says "update my marketplace", "refresh my skills", "update the claude-skills plugins", "am I on the latest skills", "install the new skill", or "why is /skillfactory still the old version". Reports installed vs available version per plugin, installs what is missing, updates what drifted, flags plugins shadowed by a stale personal copy in ~/.claude/skills, and always says whether a restart is still needed.
-
natejswenson Bundle Bible StudyCreate a researched one-page Bible study from a verse, passage, or chapter, using only the Bible and Christian sources. Use for "create a Bible study", "study John 3", or "make a Bible study handout". Includes writing date, historical context, meaning, related verses, discussion, application, and a shareable visual.
-
natejswenson Bundle GhostwriterWrite engaging LinkedIn posts in the user's own voice and publish them to their profile after they approve. Use when the user wants to draft, write, or post something to LinkedIn, asks for a "LinkedIn post", wants content about trending topics in their field, or wants to set up / configure LinkedIn auto-posting. Learns the user's voice from their past posts and never publishes without explicit approval.
-
natejswenson Bundle Github StatsShow GitHub profile statistics for a user — total commits, followers/following, stars, pull requests, and issues — as a clean summary table, plus a repository browser and per-repo detail. Use when the user asks for "github stats", "github-stats", a GitHub profile summary, "how many stars/followers/commits does <user> have", repo stats for a GitHub account, or wants to create a repo while looking at stats. Powered by the gh CLI.
-
natejswenson Bundle IssuecreatorCreate GitHub issues grounded in repository evidence and formatted for issueflow. Use when the user says "create an issue", "turn this into a GitHub issue", "write an issue for issueflow", or "break this work into issues". Supports local drafts and authorized publishing; implementing existing issues belongs to issueflow.
-
natejswenson Bundle RepocountRepocount
-
natejswenson Bundle Skillhelp 2Answer help questions about the skills in this repo. Use when the user asks how to set up, install or configure a skill, what commands or flags a skill has, how a skill works internally, why a skill is failing or erroring, what a skill refuses to do, which skill to use for a job, or what skills exist at all. Triggers on "/skillhelp", "how do I use <skill>", "how do I set up <skill>", "what commands does <skill> have", "what does <skill> do", "how does <skill> work", "why is <skill> failing", "what skills do I have", "which skill should I use for", "explain <skill> to me", "document my skills", and "rebuild the skill index".
-
natejswenson Bundle Ghostwriter 2Write engaging LinkedIn posts in the user's own voice and publish them to their profile after they approve. Use when the user wants to draft, write, or post something to LinkedIn, asks for a "LinkedIn post", wants content about trending topics in their field, or wants to set up / configure LinkedIn auto-posting. Learns the user's voice from their past posts and never publishes without explicit approval.
-
pvsundar Bundle Cgpt BackupCreate timestamped backup copies of important files or folders before editing, formatting, bulk replacement, risky automation, or user-requested backup work. Use when the user says backup, create a checkpoint, copy before editing, preserve the current version, or when a MEMORY/HANDOFF note marks a file as frozen or critical.
-
pvsundar Bundle Cgpt R DebuggerReproducible R debugging, profiling, and edge-case testing for scripts, functions, Quarto analysis chunks, and data pipelines; use when the user asks to diagnose incorrect R output, slow code, brittle behavior, vectorization options, memory use, or runtime failures.
-
pvsundar Bundle Cgpt Author ReviserExecute approved academic manuscript revisions from referee reports, decision letters, or user-approved revision plans. Use when the user wants Codex to apply reviewer-driven changes to manuscript files, but only after backup, triage, edit preview, and explicit approval.
-
pvsundar Bundle Cgpt Quarto DebuggerDiagnose and fix Quarto source problems in `.qmd` projects, especially YAML/front matter, citations, cross-references, includes, filters, and Pandoc/engine errors. Use when Quarto renders fail, output shows unresolved references, citations do not resolve, includes are missing, or a user shares a Quarto error and asks what it means.
-
pvsundar Bundle Cgpt Thread CloseoutUse when the user asks CGPT/Codex to close out, wrap up, end a thread, prepare a handoff, save session memory, create next-thread instructions, update status logs, archive a conversation, or make sure a project can be resumed cleanly later.
-
pvsundar Bundle Cgpt Manuscript RecoveryInspect an academic manuscript or revision folder after a long break and produce a durable recovery map. Use when the user asks to recover manuscript status, find the live version, identify backups/submission files/reviewer letters/Turnitin reports, resume a revision, reconstruct next actions, or create a recovery handoff before editing a manuscript.
-
pvsundar Bundle Cgpt Onedrive Git SafetyWork safely with Git repositories stored under OneDrive, Google Drive, Dropbox, iCloud Drive, or other synced Windows folders. Use before git status/add/commit/diff/reset/checkout in synced repos, when handling dirty worktrees, no-touch sibling repos, line-ending phantoms, stuck index.lock files, or closeout claims about tests/builds on synced filesystems.
-
pvsundar Bundle Cgpt Quarto Render WindowsDiagnose and harden Quarto render automation on Windows, especially from spawned PowerShell, Codex, Desktop Commander, scheduled tasks, Cowork handoffs, or CI. Use when quarto render fails because cmd/quarto/R is not on PATH, PATHEXT is stripped, artifacts are stale despite exit code 0, output paths are wrong, rBinaryPath/checkRBinary/cmd.exe spawn errors appear, or when creating reusable Windows render scripts for `.qmd` projects.
Frequently asked questions
What are Coding & Dev Tools agent skills?
Coding agent skills teach AI agents repeatable engineering workflows: reviewing pull requests, writing tests, refactoring safely, and enforcing house style. Install one SKILL.md and your agent applies the same checklist every time, whether you use Claude Code, Cursor, Codex, or another agent.
Which Coding & Dev Tools skills are most installed?
Popular Coding & Dev Tools skills on SkillMD right now include swarm, recall, unslop. Rankings shift as installs change; sort this page by "Most installs" for the live list.
Do Coding & Dev Tools skills work with Claude Code and Cursor?
Yes. Every skill here ships as a SKILL.md file, an open format that works in Claude Code, Claude.ai, Cursor, Codex, Windsurf, and 60+ other agents. Install one with npx skillmds@latest add <owner>/<name>, or copy the file into your agent's skills directory.