The library

delight · When you need one or two memorable micro-interactions that users would screenshot, not decorative animation.

clarify · When every button, error message, empty state, and form hint needs to be verb+object and actionable.

typeset · When typography is technically correct but hierarchy is invisible — need font choice, modular scale, tracking, and micro-details (tabular-nums, balance, curly quotes).

distill · When UI has redundant elements, excessive color, or competing accents that distract from primary action.

vara-skills · When building modular agent workflows where reusable skills reduce duplication and improve testability.

critique · When you need structured feedback on code quality, design consistency, or anti-patterns before shipping.

offensive-jwt · When you need to test JWT security (algorithm compliance, signature verification, expiration handling) in authorized penetration tests.

offensive-oauth · Testing OAuth implementations for authorization bypass and token leakage vulnerabilities.

traefik-route-manager · Managing dynamic routing configurations for microservices without manual Traefik config editing.

offensive-basic-exploitation · Developing and validating exploits for vulnerable software in controlled lab environments.

SKILL · Automating workflows for SKILL without manual intervention.

tdd · Enforcing test-driven development cycles to catch regressions early and guide design.

offensive-crash-analysis · Developing and validating exploits for vulnerable software in controlled lab environments.

sqlmesh · Automating workflows for sqlmesh without manual intervention.

nuxt-ui · Automating workflows for nuxt ui without manual intervention.

offensive-mitigations · Developing and validating exploits for vulnerable software in controlled lab environments.

reka-ui · Automating workflows for reka ui without manual intervention.

offensive-toctou · Developing and validating exploits for vulnerable software in controlled lab environments.

sentry-instrumentation · Automating workflows for sentry instrumentation without manual intervention.

tesserax · Automating workflows for tesserax without manual intervention.

offensive-bug-identification · Developing and validating exploits for vulnerable software in controlled lab environments.

powertools-aws · When writing Lambda functions that need structured logging, distributed tracing, and metrics in production.

ac-coverage · When you need to verify that acceptance criteria are actually tested before release.

compose-ui-control · When you need to test Compose Desktop UI automatically via HTTP without native frameworks.

commit · When you need to commit code with automatic service detection and cross-service validation.

offensive-fuzzing · When you need a methodical approach to finding memory vulnerabilities in any target.

hotfix · When you need a safe prod hotfix with guaranteed backmerge to test and main.

offensive-vuln-classes · When learning memory corruption exploit techniques with real CVE case studies.

agent-qa · When you need to validate agent behavior and confidence across multiple test cases.

validate · When you need to quickly check if all affected services pass their lint/format/test rules.

api-crud-generator · When you need a complete, tested REST API with auth and docs from a single prompt.

worktree · When you need isolated branches without stashing or checking out locally.

api-crud-generator · When you need a complete, tested REST API with auth and docs from a single prompt.

fix-lint · When you have lint errors and want automatic fixes before manual intervention.

micro-pipeline · When building any multi-step output (docs, code, reports) and you need quality gates.

create-pr · When you need to create a PR with automatic base branch detection.

what-would-cc-do · When unsure what Claude Code skill to invoke next.

offensive-shellcode · When you need robust, testable shellcode with immediate verification.

review-pr · When you need multi-angle code review (correctness, efficiency, style) with inline comments.

backmerge · When releasing a hotfix and need to synchronize prod-test-main in proper order.

offensive-windows-mitigations · When auditing Windows binary defenses or preparing Week 8 exploit bypass techniques.

offensive-iot · When assessing smart-home or ICS device security via hardware and firmware analysis.

offensive-mobile · When auditing mobile app security including deep-links WebView APIs and exported IPC.

zod · When needing runtime type checking tied to static types in form data or API payloads.

codex-cli-runtime · When delegating complex code repairs to Codex from within Claude Code subagents.

offensive-osint-methodology · When teaching or systematizing OSINT workflows to avoid ad-hoc recon drift.

skill-eval · When vetting skills for inclusion in agent harnesses or production use.

codex-result-handling · When consuming Codex outputs in automated Claude Code workflows.

offensive-osint · When starting reconnaissance on a target with only name or domain known.

SKILL · When onboarding a new skill into a framework or verifying structure.