Copy this prompt into the AI assistant you're using.
Follow https://skillsmp.com/skill-install/prompt.md to install Agent Skills from https://github.com/deadmade/dead-claude-skills.Loading the repository README…
Showing 34 of 34 collected skills.
Bump a dependency safely: read its changelog and migration guide, find every use in the code and what the breaking changes hit, approve a plan, bump, fix and test, then /verify. Resumable. Use for /upgrade <package>[@version]; no argument resumes.
WEB FRONTEND ONLY: use solely for browser UI work (HTML/CSS, React/Vue/Svelte/Astro pages and components); never for CLIs, backends, libraries, or non-web apps (e.g. Rust, Python, Go), even when asked for a new app. Anti-AI-slop design skill for greenfield…
C#/.NET only. MSBuild and build failures, NuGet packages and central package management, upgrading .NET versions, dotnet new templates. Index of Microsoft's .NET skills for these.
Use this skill for planning, testing, and safely executing PostgreSQL schema migrations — especially when working with production data or shared databases. **Trigger when user asks to:** - Test a schema migration before applying it to production - Add,…
Use for any question about a codebase, its architecture, file relationships, or project content — especially when graphify-out/ exists, where the question should be treated as a graphify query first. Turns any input (code, docs, papers, images, videos) into a…
Sketch types, signatures, and module structure before code, then stay in the loop while implementation fills in. Use for /architect, 'architect this', 'design this', or non-trivial work where jumping to code would lock in the wrong shape.
Spawn N parallel candidates at the same task, pick a base, graft the strongest parts of the losers into it. Use for /arena, 'arena this', 'throw it in the arena', or when one attempt at a non-trivial artifact would lock in the wrong shape.
Find what a change could break somewhere else before it ships, beyond the diff, and prove the one fact it's safe because of by running real code instead of writing it up. Use for 'blast radius of X', 'what could this break', or reviewing a small diff you…
Restate the last message in plain human language, with no jargon.
Use for "how does X work", code walkthroughs before changing something, and placement / ownership / layering questions ("where should this live", "which package owns this", "is this the right layer"). Explains subsystem architecture, runtime flow, onboarding…
Use for "interrogate", "adversarial review", "multi-model review", "challenge this", "stress test this code", "find blind spots", or "tear this apart". Multiple LLM reviewers challenge changes from independent angles.
Spawn Comment Sicko, fix accepted findings, and offer encodings for claimed constraints.
Apply when wiring validation, error handling, or framework adapters. Concentrate guards at system boundaries (CLI, config, network, external APIs); trust internal types and keep business logic in pure functions.
Apply when designing commands, lifecycle steps, or processing loops that run amid crashes, restarts, and retries. Converge to the same end state regardless of partial prior runs.
Apply when writing stateful logic, or when code branches a lot or repeats a shape assumption across files. Encode the domain in a structure instead of scattered conditionals.
Apply when integrating a new requirement into an existing design. Redesign as if the requirement had been a foundational assumption from day one, instead of bolting it on.
Apply when concurrent actors might write to the same file, branch, key, or state object. Eliminate the sharing first; serialize structurally only when one shared writer is a real invariant.
Apply when designing types, reviewing a function signature, or writing code in any statically-typed language. Make illegal states unrepresentable, brand semantic primitives, parse external data at boundaries, refuse to lie to the compiler, exhaust variants,…
Reconstruct your recent working context from your own chat history, live state, and the shared record (user reports, prior fixes, incidents), then hand back a tight current-state brief. Use for 'recall my work on X', 'catch me up', 'what have I been working…
Keep a reviewable decision trail for long-running or unattended work: a TSV log with one row per decision (what, why, evidence, result). Local by default; commit it when a reviewer needs the trail to trust the result. Use for /show-me-your-work, autonomous or…
Layered technical-writing standard: Diátaxis structure, Google developer style sentences, STE instruction rules, Global English syntax. Use for /technical-writing or when writing or reviewing docs, RFCs, readmes, PR descriptions, or commit messages.
Cut AI tells from any writing. Must always apply.
Use for 'why does X work this way', 'why we picked Y', design rationale, regressions, postmortems, or data-backed thresholds. Discovers available MCPs and queries each evidence category (source control, issue tracker, long-form docs, real-time chat,…
Review your own branch before opening a PR: a confidence-scored bug pass, a spec pass, a standards pass, a security pass, a CLAUDE.md audit, and specialist agents for error handling, tests, types and comments, all in parallel, then one ranked report and fixes…
Check that the docs tell the truth about the code: every claim in READMEs, docs/, CLAUDE.md and manifest descriptions held against the code (contradicted, unbacked, undocumented), commands, paths, links and versions checked, changed prose edited, then one…
Build a feature through the whole skill chain: ground in the code and in how other projects solve it, brainstorm with options, grill the spec with domain docs and ADRs, write the plan, evaluate it adversarially, then execute, verify and finish. Two approval…
Fix a bug from symptom to verified branch: reproduce it, map where it lives and what a fix could break, search upstream for the same issue, find the root cause with systematic debugging, approve it, write the failing regression test first, fix, then /verify…
Find what makes the code fragile: security holes, footgun APIs and dangerous defaults, bloat and dead code, and shallow modules worth deepening, from four passes run in parallel, merged into one ranked report, then fixes for the findings you pick. Use for…
Build from a plan, spec or ticket that already exists, without the design phases of /feature-flow: turn a spec into tickets or a plan if needed, approve the plan, execute it test-first through superpowers, then /verify and /learn. One approval gate,…
Keep what a piece of work taught, so the next session starts smarter: collect the surprises, decisions and root causes from a flow or the current session, sort each to its home (CLAUDE.md, CONTEXT.md, an ADR, docs/solutions/, or memory), and write the ones…
Build a feature on scientific evidence: ground it in the code and in papers, brainstorm options that cite an evidence table, state a falsifiable hypothesis with a metric, grill, plan, evaluate, execute, then measure the result against the hypothesis before…
Is the test suite appropriate? Run it, then find what it would miss: mutation testing (surviving mutants are weak tests), property-based testing where the code is a parser, codec or invariant, requirements with no test, and test smells. One ranked report,…
Learn a codebase, a subsystem or a concept by doing: map it (graphify, pstack how), learn why it is that way (pstack why), then a guided lesson through mattpocock teach where you write the key lines and Claude explains, and keep what you learned in…
Prove a branch is ready for a PR: run the tests and checks and read their output, /self-review the diff, /doc-audit the docs it touched, optionally /test-audit it, read it in tui-review, then give a verdict and offer how to integrate. Never opens a PR or…
The repository README could not be loaded. You can still explore its Skills below.