Estimate how long a dev task will really take when it's built AI-assisted (an agent writes the code, a human reviews, tests, ships). Grounds every number in the actual code first, counts agent work in tool-call rounds, then adds the human layer (diff review, local run, browser QA, PR/CI, review feedback, PM back-and-forth) as separate lines. Use this whenever someone asks for an estimate, "how long would X take", "how many hours", sizing, scoping, a quote for a PM or client, or a comparison of options by effort — even when they don't say "estimate" and even for a quick ballpark. Also use it before replying to a ticket, Lobby/Teamwork/Slack thread or PR comment that asks "estimate to complete?". Adapted from ZhangHanDong/agent-estimation (MIT).
1 skill
App Store Optimization for iOS and Android from one per-locale Markdown source.
1 skill
Render audit findings as a designed, paginated A4 PDF report with a cover page, severity donut and category bar charts, colored severity chips, evidence-backed strengths, prioritized recommendations, and copy-ready GitHub issue blocks.
1 skill
Babysit a GitHub PR from review all the way to MERGED.
1 skill
Turn a settled scope into an implementation plan at docs/plans/NNNN-<slug>.md: files affected, phased sequence gated on numbered acceptance examples, edge-case ledger, rollback and risk. Consumes PRDs + ADRs. Formerly heist.
1 skill
Maintain a monthly brag document and roll it up into a promo packet, self-review, CV material or interview story seeds. Lucas Faria's 4-field template + Julia Evans' impact-first categories, with ownership, status, evidence and story seeds recorded each month. Feeds cv-craft.
1 skill
Build a one-page business brief on a company the user is about to work for or with, so they walk in knowing how it makes money and where their work moves the number.
1 skill
Apply Marty Cagan (SVPG) + Teresa Torres Continuous Discovery to a dev's workflow: feature kickoff, planning/estimation, discovery calls, PR review. Flags feature-factory smells and business-sense gaps (P&L chain, business-model lever, prize vs. build hours); green/yellow/red per dimension.
1 skill
Language- and framework-specific best practices and idioms. Reader and reviewer modes. Ships 7 languages + 12 frameworks (TS, Rust, CSS, Tailwind, Dart, Go, Python, React, Next, RN+Expo, Hono, Workers, Astro, Svelte, Drizzle, Supabase, Lambda, Terraform, Flutter).
1 skill
Work report in two modes — quick (git-only: prose + bullets from the current repo) and standup (multi-source: GitHub PRs/reviews/comments, git across repos, deploys) — tunable to audience (dev / pm / client), delivered as a copy-ready channel block or a doc file.
1 skill
Multi-mode software engineering coach: architect, advisor, reviewer, refactor, legacy, and explainer workflows. Auto-writes ADRs to docs/adr/.
1 skill
Maintain a canonical master CV in Markdown and produce tailored, ATS-friendly resumes, recruiter-screen prep packs and an evidence-backed interview story bank.
1 skill
Conductor for the dev pipeline: detects a task's tier (bug / feature / architecture / client), prints the exact chain of sub-skills it will run and skip with reasons, confirms, then drives them in order. Matches ceremony to stakes.
1 skill
Stress-test a plan or design via relentless interview. Resolves each branch of the decision tree, captures domain vocabulary and a concrete example per business rule, emits Design Notes + Glossary block for /to-prd. Adapted from mattpocock/skills.
1 skill
Pre-commit quality gate: seven checks (plan gaps, use-case coverage with AC traceability, test scenarios, test philosophy and integrity, SOLID, Clean Code, security) run in fresh subagents, seven in parallel or one for a small diff, plus a tests-only mode. Runs after /run, before commit.
1 skill
Build and maintain an LLM-maintained wiki — a persistent, interlinked Markdown knowledge base distilled from curated raw sources (Karpathy's LLM Wiki pattern).
1 skill
Playbook for parallel work across git worktrees: go/no-go, file partitioning, isolation mechanism, integration and cleanup. Works with or without formal planning docs.
1 skill
Role-play a senior, staff or principal engineer and sweep the surfaces you declare for tech, product and org opportunities. Returns at most 5 ranked openings, each with a concrete first move and a draft. Read-only until you approve a specific message.
1 skill
Sync the current project into a central project-portfolio Markdown archive. Source data for cv-craft.
1 skill
Create a pull request with a structured description (Problem / Root cause / Fix / Test / Out of scope) derived from the diff and conversation. Drives the branch -> commit -> push -> gh pr create flow.
1 skill
Problem-solving coach for real decisions and messy problems that are not about code.
1 skill
Publish one of your own skills from ~/.claude/skills to this repo: copy it under skills/<name>/, mint its plugin.json, upsert marketplace.json, bump SemVer, and open a PR (CI-gated). Idempotent.
1 skill
Single review-only entry point for an existing diff — runs verify → code-review → implementation-review (and security-audit only when the change touches a trust boundary), then consolidates every finding into one go/no-go verdict.
1 skill
High-confidence security review of a diff, layered on wstg-security-testing. Confidence gate, false-positive precedents, WSTG-ID mapping, and a dependency audit. Reports only concretely exploitable findings.
1 skill
Decide the correct Semantic Version bump for a change (major/minor/patch) by analyzing the diff against the project's public surface and running the ecosystem's breaking-change detector, then apply it (bump the manifest + changelog).
1 skill
Close the loop after ship — scan a repo (and optionally its live prod signal) for post-release trouble, then file what it finds as docs/intent/NNNN-<slug>.md so it re-enters the dev-flow chain as ordinary work.
1 skill
Deterministic LLM-first SEO audits for websites, blog posts, and GitHub repos. 16 sub-skills, 10 specialist agents, 88 evidence scripts. Adapted from Bhanunamikaze/Agentic-SEO-Skill (originally AgriciDaniel/claude-seo).
1 skill
Defines what a good test is: behavior over implementation, the Testing Trophy, an e2e floor for user-facing features, acceptance tests as Given/When/Then examples, and integrity rules for tests an agent writes or changes. Stack-agnostic (TS/Rust/Go/Elixir/Python). Kent Dodds, generalized.
1 skill
Synthesize conversation context into a PRD. Writes docs/prds/NNNN-<slug>.md + publishes to GitHub Issues. Includes a Glossary and numbered acceptance examples (rules plus Given/When/Then). Adapted from mattpocock/skills.
1 skill
Meta-skill: create new Claude Code skills with proper structure, progressive disclosure, and bundled resources.
1 skill
Web application security testing via the OWASP Web Security Testing Guide (WSTG) — 12 categories, ~109 test cases. Four modes: guide a pentest, self-review your app, build/score a coverage checklist, review a diff against WSTG. Mode 2 carries a systematic whole-codebase audit protocol. Methodology from OWASP/wstg. Authorized/defensive use only.
1 skill