Skill Claude CodeCodex
Find the mistakes agents keep repeating in a repo and make each one impossible, trying architecture first, then types, lint, tests, and docs last. Use when the operator keeps correcting the same mistake, or for /agents-correct.
Skill Claude CodeCodex
Generate a project-local verify skill that drives the app like a user through a browser with Playwright, a CLI, or HTTP. Use for /agents-create-verification-skill, "make a verify skill for this repo", or when a project has no scripted way to prove UI, CLI, or service behavior.
Skill Claude CodeCodex
After a work session, review what went wrong and right and turn each lesson into a concrete edit to an existing skill or AGENTS.md line. Use when the user says "reflect" or /agents-reflect.
Skill Claude CodeCodex
Fan out N parallel subagent workers in one turn, aggregate their results, and report PASS, ISSUES, or BLOCKED per worker. Use for /agents-swarm, "swarm this", parallel coverage across slices, races between approaches, or broad exploration.
Skill Claude CodeCodex
Change, plan, deploy, or debug infrastructure declared with Alchemy (v2) and secrets from Infisical. Use for alchemy stack edits, bindings, Durable Object classes, D1 or other migrations, custom domains, local Docker stacks, adding or rotating a Worker secret, secret drift between Infisical, CI, and the Worker, deploy…
Skill Claude CodeCodex
Explain how a subsystem works, its architecture, runtime flow, and where new code belongs. Use for "how does X work", walkthroughs before changing something, and placement or ownership questions like "where should this live" or "which package owns this". Use comms-why for motivation.
Skill Claude CodeCodex
Multi-reviewer critique of a diff, run as parallel read-only subagents from distinct angles, then synthesized into one verdict. Use for "interrogate", "adversarial review", "challenge this", "stress test this code", "find blind spots", or "tear this apart".
Skill Claude CodeCodex
Layered technical-writing standard covering document modes, sentence style, instruction rules, and unambiguous syntax. Use when writing or reviewing skills, docs, READMEs, RFCs, PR descriptions, or commit messages.
Skill Claude CodeCodex
Cut machine-written patterns from prose while keeping meaning and voice, or flag them without rewriting. Use when drafting or editing prose, docs, PR descriptions, commit messages, or replies that read machine-written.
Skill Claude CodeCodex
Explain why code is shaped the way it is, using git history, PRs, and connected MCP evidence, with citations and honest confidence. Use for "why does X work this way", "why did we pick Y", design rationale, regressions, and postmortems. Use comms-how for runtime behavior.
Skill Claude CodeCodex
Decode untrusted data once at ingress with Effect Schema (v4) and pass typed values downstream. Use for HTTP bodies, params and query strings, webhooks, RPC and message payloads, MCP tool arguments, env and config, JSON.parse, stored JSON or rows, third-party API responses, unsafe casts, or porting Zod/Valibot schemas.
Skill Claude CodeCodex
Move heavy or long-running work off a constrained runtime (for example the Cloudflare Workers memory or CPU ceiling) by first extracting domain logic into packages as Effect programs, then moving execution. Use when starting or continuing a runtime migration or moving a service across app boundaries.
Skill Claude CodeCodex
Apply Effect v4 conventions for services, layers, tagged errors, boundary runners, resources, retries, deadlines, HTTP, and tests. Use when writing or reviewing TypeScript that imports effect, adds a service or error class, chooses where to run a program, bounds external I/O, or tests Effect code.
Skill Claude CodeCodex
A scoring scale for evaluating how well a CLI is designed for AI agents, based on the "Rewrite Your CLI for AI Agents" principles. Use when reviewing, designing, or comparing a CLI for agent use, or scoring its agent readiness.
Skill Claude CodeCodex
Vet a performance measurement for limiter, tuning, limits, errors, repeatability, relevance, and whether the work happened before you report or act on it. Use when you run a benchmark or report a measured speedup or regression.
Skill Claude CodeCodex
Find what a change could break somewhere else before it ships, beyond the diff, and prove the one fact it is safe because of by running real code. Use when asked for the "blast radius of X" or "what could this break", or when reviewing a small diff you do not trust.
Skill Claude CodeCodex
Short checklist of engineering principles to apply before and after non-trivial work. Use when planning or reviewing a non-trivial change, debugging a stubborn failure, designing types or boundaries, or deciding how to sequence a migration.
Skill Claude CodeCodex
Prove a migration, module split, library swap, datastore or deploy-tool cutover, compatibility removal, or multi-PR cleanup is complete. Use when retiring paths, adapters, dependencies, generated outputs, or config, validating merged main, or proving a rebase preserved a patch series.
Skill Claude CodeCodex
Fix a bug test-first when there is a clear, cheap regression test path. Use when the user asks for TDD, a failing test, or a regression test, or when the bug has an obvious cheap local test target.
Skill Claude CodeCodex
Diagnose failing GitHub Actions CI from log evidence and choose a fix, rerun, merge-forward, or escalation. Use when a PR or default-branch run is red, cancelled, or flaky, or a visual baseline or artifact is missing.
Skill Claude CodeCodex
Review new commits on open GitHub PRs against their actual base without stale or duplicate findings. Use for cutoff-based or scheduled review sweeps, stacked or rebased PRs, cleaning up review threads, or posting actionable inline findings. Read-only except comments and thread resolution.
Skill Claude CodeCodex
Publish a GitHub pull request and carry it through CI, review feedback, rebases, and merge readiness, merging only with explicit authorization. Use when opening or updating a PR, addressing review comments, monitoring checks, rebasing, or continuing a PR to a terminal state. For read-only review of new commits, use…
Skill Claude CodeCodex
Triage Sentry errors for one environment, trace a finding to current code, reproduce it locally, and deliver a focused fix PR. Use when sweeping an environment's errors, decomposing catch-all log or 5xx issues, or driving a supplied Sentry issue to a fix. Use sentry-perf for latency and gh-ci-triage for CI.
Skill Claude CodeCodex
Find and fix performance regressions from Sentry spans, transactions, dashboards, and web vitals. Use when asked for a performance dashboard summary, slow routes, week-over-week regressions, comparable measurement windows, or a perf fix PR. Use sentry-bugfix for errors and gh-ci-triage for CI.
At most 3 mods per repository are shown here, and a mod shipped inside a plugin is left to that plugin's page — the rest are on their repository pages: