A static catalog of Homebrew tools, Neovim plugins, agent skills, and public template repositories.
Last synced:
Last synced:
Last synced:
Use this skill when the user explicitly asks to map, document, or onboard into an existing codebase. Trigger for prompts like \"map this codebase\", \"document this architecture\", \"onboard me to this repo\", or \"create codebase docs\". Do not trigger for routine feature implementation, bug fixes, or narrow code edits unless the user asks for repository-level discovery.
Use this skill when building AI agents that call external tools, implementing policy-based access controls, adding semantic intent classification, creating trust scoring systems, building audit trails, or enforcing rate limits and content filters on agents.
Use when creating or updating agent instruction files (AGENTS.md for Pi, copilot-instructions.md for Copilot, per-path guides, or AGENTS.md router) — especially when instruction files are too long, generic, or stale, or when agents repeatedly make the same avoidable mistakes.
Use this skill when generating SHA-256 integrity manifests for agent plugins, verifying that installed plugins match their manifests, detecting tampered files, auditing dependency pinning, or building provenance chains for plugin promotion.
Use when designing or implementing an evaluation loop for AI agent outputs — reflection loops, evaluator-optimizer pipelines, LLM-as-judge scoring, or rubric-based iteration. Not when running an existing test suite or reviewing a completed artifact without iterating.
Run quick, repeatable API smoke tests after changes — verify auth, status codes, and response shapes with hurl. Use for smoke testing, sanity-checking endpoints, or confirming API health before broader testing.
Use when you need ast-grep structural search, linting, or safe codemod rewrites, especially for requests that mention ast-grep, sg, codemod, rule files, or syntax-aware edits instead of plain-text search.
Audit and improve SEO for Astro sites, including metadata, structured data, sitemaps, indexing, Open Graph images, schema endpoints, and related search visibility setup.
Use when running autonomous iterative experiments to optimize a measurable metric on an existing codebase — not for writing tests or diagnosing a specific failure.
Use when deploying or debugging Go-based AWS Lambda packaging, bootstrap, runtime, or handler issues — not when the main failure is SAM stack wiring, IAM OIDC auth, or a generic GitHub Actions run failure.
Use when the codebase still imports from 'aws-sdk' v2, uses new AWS.*() constructors, .promise() calls, AWS.config.update(), DocumentClient, or s3.upload() — and needs a safe modular v3 migration without changing runtime behavior.
Validate CI checks locally and manage Docker image publishing workflows. Use when changing build inputs (generated code, formatting, tests), preparing a release, or verifying that image tags are correct.
Use when a repository still depends on .circleci/config.yml jobs, workflows, or artifacts that need to be replaced or phased out with GitHub Actions — including incremental shadow runs, staged cutover, and parity checks.
Create, inspect, apply, baseline, import, or recover Cloudflare D1 schema migrations across local, preview, and remote databases. Use when Wrangler D1 migration state or schema drift is the main risk.
Tune slow Cloudflare D1 queries and D1-backed repository access without schema or migration changes. Use for N+1 access, over-fetching, pagination, or runtime D1 query-shape issues.
Run and troubleshoot repository code generation after detecting the generator contract. Use when schema, query, spec, or template inputs changed, or generated output may be stale.
Use when navigating or refactoring code — choosing between LSP, rg, or semantic search tools; an LSP call returns empty and you need to know if it failed; or a tool substitution needs disclosure.
Use when asked to create a .tour walkthrough — onboarding, PR review, RCA, architecture, or any persona-targeted step-by-step code narrative. Not for codebase exploration without a .tour file output.
Configure applications safely through environment variables, .env files, and repo-owned runtime config such as Wrangler bindings. Use for startup failures, environment drift, or deployment bootstrap.
Use when the user explicitly wants a pre-edit scoping pass — mapping candidate files, tests, and reference patterns before implementation. Not for general multi-file tasks unless the user asks "what files are involved" or requests a context map.
Create, apply, inspect, and recover database schema migrations after detecting the repository's migration contract. Use when changing persistent schema, adding indexes/constraints, or troubleshooting migration state.
Use when user wants to write, refactor, or expand documentation through iterative co-authoring with reader feedback — README, guides, API docs, runbooks, specification documents. Not for solo writing, readability audits, or metadata review (use writing-and-editing for those).
Run a service stack locally using Docker Compose for prod-like development (e.g. PostgreSQL + app). Use when you need a real DB backend, multi-service integration, or want to reproduce production-like behavior locally.
Use when auditing JS/TS code health with Fallow - dead code, duplication, boundaries, or cleanup; not for debugging failures.
Helps users discover and install agent skills when they ask questions like "how do I do X", "find a skill for X", "is there a skill that can...", or express interest in extending capabilities. This skill should be used when the user is looking for functionality that might exist as an installable skill.
Use when implementation work is complete and you need to decide how to integrate the branch — merging locally, creating a PR, keeping the branch, or discarding the work.
Use when commits, tags, or pushes are blocked by GPG or SSH signing failures — including 1Password op-ssh-sign errors, failed to fill whole buffer, or misleading op whoami output, not when the main task is branch integration, PR handling, or secret-scan triage.
Set up and manage isolated git worktrees for parallel tasks, agent lanes, and safer branch isolation. Use when worktree setup or recovery is the real need, not generic planning or implementation guidance.
Diagnose and minimally fix failing GitHub Actions runs in repositories already using GitHub Actions.
Use when a GitHub Actions failure needs local reproduction with act before pushing, especially for workflow logic or step-level regressions, but not when the failure depends on org-managed runner or secret policy.
Use when a branch needs GitHub CLI PR lifecycle work, like creating or updating a PR, watching checks for a pushed head SHA, or preparing PR handoff before review.
Audit and improve GitHub profile, organization, and repository presentation. Use for profile READMEs, pinned repositories, README quality, community health files, templates, and repo metadata.
Diagnose Go build failures, toolchain mismatches, and CI-parity problems — GOROOT/GOTOOLDIR overrides, govulncheck stdlib findings, or TempDir race flakes. Use when go build or CI fails for environmental reasons, not when individual tests fail on logic or coverage (use testing-workflows for that).
Develop, structure, and release Go CLI tools. Use when building a new Go CLI, adding subcommands or flags, wiring config, formatting output, setting up goreleaser + Homebrew distribution, or debugging CLI-specific behavior — not for build failures or test debugging.
Containerize Go services with production-ready multi-stage Docker builds. Use when writing or debugging a Dockerfile for a Go service, setting up static compilation for scratch/distroless images, adding health checks, or wiring image builds into CI — not for Docker Compose local dev stacks or non-Go Dockerfiles.
Use when Go error-handling design or function-type mismatches surface — including `func(string) error` vs concrete constructors, package-local type aliases, or concrete-vs-interface error choices, but not when the immediate task is a build or test failure.
Use when setting up, debugging, or fixing a GoReleaser v2 release pipeline — especially Release Please tag issues, missing release assets, or publish jobs that fail after a release is created, not when the main task is generic CI triage or PR handling.
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 persistent knowledge graph with god nodes, community detection, and query/path/explain tools.
Use when the user wants to be interviewed or stress-tested about a plan, design, or decision — or when the user explicitly says 'grill me'. Not for domain-doc maintenance; route to grill-with-docs when CONTEXT.md or ADRs should be updated.
Use when stress-testing a plan against the project's domain model, when domain terminology is fuzzy or inconsistent, or when the user says 'grill me' about a design that should produce or update CONTEXT.md and ADRs.
Keep OpenAPI contracts in sync with HTTP behavior. Use when you need to update spec files, code annotations, or both after changing endpoints, schemas, auth, or error shapes — not for writing or running integration tests.
Use when GitHub Actions or another OIDC caller gets AWS sts:AssumeRole or sts:AssumeRoleWithWebIdentity AccessDenied — not when auth succeeds and the remaining failure is in SAM, CloudFormation, or Terraform execution.
Review completed code changes, diffs, and implementation revisions for merge readiness. Use for code review, named reviewer approval, validation gaps, regressions, security risk, or scope drift.
Write and run end-to-end HTTP integration tests for server behavior. Use when you need to add or fix integration test coverage for handlers, middleware, auth enforcement, or endpoint contracts — not for keeping OpenAPI specs in sync.
Use when a JavaScript or TypeScript project needs Knip to find unused dependencies, exports, files, or unresolved imports; not for runtime failures or dormant-task recovery.
Polish existing interfaces when the UI feels off through better radius, shadows, typography, motion, and micro-interactions.
Use when a JavaScript or TypeScript suite still depends on Mocha, Chai, or Sinon and needs an incremental Jest migration.
Consult this reference before implementing any HTML, CSS, JS, form, or animation feature — prevents legacy patterns by checking for modern standardised web platform APIs. Load as a reference overlay, not as a primary workflow skill.
Use when editing, debugging, or validating a Neovim Lua configuration — including plugin management (vim.pack, lazy.nvim), LSP/completion wiring, treesitter, statusline, or startup behavior.
Develop, test, and release Neovim Lua plugins. Use when building a new Neovim plugin, adding features to an existing plugin, setting up CI with GitHub Actions, writing plenary tests, generating documentation, or configuring release automation — not for editing user-level Neovim config.
Validate and operate application /health, /metrics, and logging behavior. Use when adding metrics, changing observability endpoint auth, or verifying deployment telemetry — not for general production debugging where telemetry health is not the question.
Create, revise, and approval-gate implementation plans when the deliverable is a plan artifact or plan revision, not code.
Use when a completed /plan needs explicit multi-reviewer approval and refinement before implementation begins.
Use when a layered TypeScript workspace needs an incremental project-references migration without breaking package boundaries or editor resolution.
Implement or modify database repository adapters. Use when adding adapter methods, changing queries, or mapping database errors while preserving domain boundaries.
Use when the stabilisation-guard blocks the session because of active open_loop or assistant_goal memories, or when the user explicitly asks to review and close out pending items in Lore.
Rewrite a rough request into a repository-grounded execution brief before research, planning, or implementation. Use when the user asks for a sharper prompt, better request, or contract-shaped brief.
Apply pull request review feedback by assessing comments, fixing valid concerns, pushing the branch, and waiting for checks. Use when reviewer feedback must be carried through to completion, not just reviewed.
Apply a Research → Plan → Implement → Validate workflow for multi-phase work that benefits from explicit phase boundaries. Use when the task is complex, spans many files, or needs durable progress artifacts — not for small fixes, single-file edits, or narrow specialist workflows.
Use when AWS SAM templates, CloudFormation stacks, or SAM deploys fail on YAML, transforms, resource wiring, or rollback state — not when the Lambda binary itself or IAM OIDC auth is the primary blocker.
Introduce or refine runtime schema validation at untrusted boundaries so static TypeScript types stay truthful.
Use when secret scanning with gitleaks reports findings that need fast triage, containment, and false-positive adjudication before code changes or merges continue.
Apply security and privacy guardrails to application code. Use when touching authentication, secrets, sensitive endpoint exposure, or logging of credentials/PII — not for general request-handling changes without a security dimension.
Search and summarize past work from the `session_store` database when the user asks what they worked on before, how a similar problem was handled, or which session, PR, issue, branch, or file changed something.
Use when designing the shape of a new or revised skill — deciding activation language, layering, examples, and validation patterns. Consult this as the authoring style reference, not as a step-by-step builder. For building or upgrading a specific skill package, use skill-creator instead.
Build, upgrade, or standardise a specific skill package in the current catalog. Use when adding a new skill directory, improving its metadata/examples/evals, or tightening its trigger boundaries — not for general skill design philosophy or style reference (use skill-authoring for that).
Use when encountering any bug, test failure, unexpected behavior, or persistent error before proposing fixes — especially under time pressure, after multiple failed attempts, or when the root cause is unclear.
Edit and troubleshoot server-side HTML templates (.templ files for Go's templ framework), including regeneration and handler wiring. Use when changing UI components, layouts, or page templates.
Use when writing, reviewing, or debugging Terraform/OpenTofu — modules, tests, CI/CD, security scans, or state operations.
Use when the user explicitly asks for test-driven development (TDD), a red-green-refactor cycle, or test-first implementation — not for general feature work, bug fixes, or refactoring unless TDD discipline is specifically requested.
Write, run, and debug Go tests — unit and integration — including generator refresh when stale generated code causes failures. Use when adding test coverage or debugging test failures in domain logic, repositories, or handlers. For build failures, toolchain issues, or CI-parity problems, use go-build-and-test instead.
Use when the user wants to turn a plan, spec, PRD, or approved idea into independently grabbable issue-tracker slices. Not when the work still needs a PRD, repository discovery, or a generic handoff artifact.
Use when the user wants a PRD synthesized from current repository and conversation context, especially for issue-tracker handoff. Not when doc-coauthoring, workflow-contracts, or codebase discovery is the better fit.
Diagnose and resolve TypeScript compiler errors by finding the smallest root-cause fix before editing call sites.
Tighten and rationalize TypeScript configuration safely, especially when enabling stricter compiler checks incrementally.
Write compile-time type tests that protect public TypeScript APIs, inference behavior, and negative cases.
Use when TypeScript source contains explicit `any` in application code, shared helpers, DTOs, or API layers that should be narrowed without changing runtime behavior — or when a reviewer or lint rule flags unsafe `any` usage.
Use when about to claim work is complete, tests pass, a bug is fixed, or a build succeeds.
Use when a task needs a durable handoff artifact for planning, review, or execution in agent-driven workflows — not just a chat summary.
Use when a user needs to configure or operate Worktrunk (wt) for worktree lifecycle management, hooks, LLM commits, or parallel agent workflows.
Draft structured documents, audit prose readability, and review short audience-facing metadata strings — solo writing quality work. Use when the primary task is writing, editing, or copy quality, not iterative co-authoring with reader feedback (use doc-coauthoring for that).
Last synced:
Astro starter template for building and deploying sites to Cloudflare Pages.
Starter CRUD API template with Elysia, Drizzle ORM, Postgres, and Bun.
Minimal Go API template with Chi, Docker, CI, and GHCR publishing.
Template repository for building and releasing Neovim plugins with tests and CI.
TypeScript package template with dual CJS/ESM builds, testing, linting, and CI.