The verified registry for AI coding agents: security-checked plugins and portable SKILL.md skills. Open eight-check methodology, machine-readable results, consent-first curation.
Agent
Use this agent when accessibility findings need to be FIXED, not just reported, connecting a control to its label across the DOM, choosing the correct ARIA pattern from surrounding structure, mapping a failing color to the nearest accessible design token, or fixing keyboard focus management. This is the cross-element…
Agent
Use this agent when hardcoded user-facing strings need to be extracted into a translation catalog and their call sites rewritten with the framework's idiomatic translation call, preserving interpolation and pluralization as correct ICU. Trigger phrases include "extract these strings", "internationalize this…
Agent
Use this agent when an OpenAPI/Swagger or GraphQL spec has drifted from the handler code and needs to be reconciled, an undocumented endpoint, a param the code accepts but the spec omits, a response field the code returns that the schema does not declare, a status code or nullability mismatch, or a spec that describes…
Agent
Use this agent when background-job or message-consumer code needs an at-least-once reliability audit, a worker with no dead-letter queue, no idempotency/dedup store, no retry backoff, an ack before the work is done, or an event published without a transactional outbox. Trigger phrases include "audit my queue", "are my…
Agent
Use this agent when a CI pipeline is failing and you need the root cause and exact fix, across GitHub Actions, GitLab CI, or CircleCI. Trigger on "CI is red", "workflow failed", "pipeline broken", "why did the build fail", "actions permission denied", "secret not found", or a pasted CI log. Examples.
Agent
Use this agent when a Dockerfile should be made smaller, safer, and faster, multi-stage build, non-root user, pinned base, cache-friendly ordering, and a matching .dockerignore. Trigger on "harden my Dockerfile", "reduce image size", "container runs as root", "add multi-stage build", or a scanner flagging an image.…
Agent
Use this agent when an incident is resolved and you need a blameless postmortem assembled from raw evidence. Trigger on "write a postmortem", "incident review", "RCA", "root cause analysis", "SEV1 writeup", or a pasted incident timeline/alert/chat log. Examples.
Agent
Use this agent when the user asks about system architecture, component structure, dependency graphs, design patterns, module coupling, or wants architecture diagrams generated. Examples.
Agent
Use this agent when the user asks about code quality, code smells, complexity analysis, maintainability, code standards, duplication, or wants a code review focused on quality. Examples.
Agent
Use this agent when the user asks about performance, optimization, bottlenecks, bundle size, caching strategy, runtime efficiency, memory leaks, or rendering performance. Examples.
Agent
Use this agent when the user discusses security vulnerabilities, asks for security review, mentions hardcoded secrets, injection attacks, authentication issues, or OWASP compliance. Examples.
Agent
Use this agent when the user asks about test coverage, test quality, missing tests, testing strategy, test-to-code ratio, or wants to evaluate their testing approach. Examples.
Agent
Use this agent when someone knows WHAT to change but not WHERE in the codebase to change it, task-scoped localization that returns the precise files, functions, and tests to edit. Trigger on "where do I add ...", "which file handles ...", "where is X implemented", "what do I need to touch to ...", "find the code for…
Agent
Use this agent when a repository won't run locally and you need it bootstrapped and green, detecting the required runtime/tooling versions and services, running the setup, and diagnosing failures. Trigger on "get this running", "set up the dev environment", "the install is failing", "wrong Node/Python/Ruby/JDK…
Agent
Use this agent when you need to understand what actually happens end-to-end when a request or user action runs, following routing, middleware, handlers, services, and data access across the real call graph. Trigger on "trace this request", "what happens when a user clicks X", "walk me through the login flow", "how…