Agent Claude Code
Deep, sourced web-research engine. Confirms every datum with ≥2 independent sources, extracts verbatim quotes, flags contradictions and unknowns. Output: findings+provenance JSON artifact (written to file) + one-line return. Shared worker for ds-research and ds-brief.
Skill Claude CodeCodex
Backend architecture — API design, database schema, authentication, data pipelines. Use when designing or reviewing a backend, REST/GraphQL APIs, data models, auth flows, or ingest/ETL pipelines.
Skill Claude CodeCodex
Ideal-vs-current gap analysis — research 5-10 comparable projects, synthesize the ideal architecture, produce a dimension-by-dimension gap table. Use when benchmarking against best-in-class or deciding which architectural gaps to close.
Skill Claude CodeCodex
Project health system — profile-based assessment, transformation, and progress tracking. Use when assessing project health, planning a transformation, or tracking progress over time.
Skill Claude CodeCodex
Data-backed brief — research, source, double-verify, and render a visually rich single-file HTML report. Use when the user wants a sourced, fact-checked brief or report on a topic.
Skill Claude CodeCodex
Smart commits — quality gates, atomic grouping, Conventional Commits formatting. Use when committing changes, grouping a working tree into atomic commits, or writing commit messages.
Skill Claude CodeCodex
Security and regulatory compliance — OWASP, privacy laws, data protection, web security, i18n. Use when auditing for security/privacy compliance, GDPR/KVKK, or pre-release legal review.
Skill Claude CodeCodex
Deployment and infrastructure — containerization, deployment, monitoring, incident response. Use when deploying an app, setting up containers/monitoring, or planning incident response.
Skill Claude CodeCodex
Dependency upgrade loop — classify each dependency as safe-patch/safe-minor/review-major, apply safe groups with per-group commits, surface majors with migration notes and rollback. Use when upgrading dependencies or auditing for outdated/vulnerable packages.
Skill Claude CodeCodex
DevOps audit — CI/CD pipelines, code signing, dependency management, deployment config for any stack. Use when auditing or setting up CI/CD and release infrastructure.
Skill Claude CodeCodex
Documentation integrity — gap analysis, doc-drift verification, ADR tracking, and AI-harness context-file curation. Use when docs are incomplete or stale, the user asks to write/improve README or API docs, wants a decision recorded, or wants CLAUDE.md/AGENTS.md trimmed.
Skill Claude CodeCodex
Universal code quality fix — format, lint, type-check, l10n, and security scan for any stack. Use when the user asks to fix code quality, run formatters/linters/type-checks, or scan for secrets.
Skill Claude CodeCodex
Release scope triage — decide which features ship this release vs defer to backlog, then file/implement the kept set and sync docs to match. Use when a release has grown too complex to finish or polish everything and the team needs to cut scope deliberately.
Skill Claude CodeCodex
Frontend design quality — design-system audit, token enforcement, component states, accessibility, responsive layout, theming. Use when building or reviewing UI for design consistency, a11y, or responsiveness.
Skill Claude CodeCodex
Project scaffolding — generate production-ready project structure for any stack. Use when starting a new project or bootstrapping a repo skeleton.
Skill Claude CodeCodex
GitHub-Issues lifecycle in one skill — file a verified deduped issue, sweep the set for duplicates, audit done-ness from code, and execute an issue end-to-end (re-verify → impact map → implement → code-proven close). Use to open, refine, sweep, status-check, or do an issue.
Skill Claude CodeCodex
Store and release management — store submission, listing optimization, release strategy, post-launch monitoring. Use when preparing an app-store release or planning a launch.
Skill Claude CodeCodex
Mobile app quality audit — 181 rules across 13 domains with release-readiness scoring (Flutter, SwiftUI, Kotlin/Compose, React Native, Capacitor). Use when auditing a mobile app for quality or release readiness.
Skill Claude CodeCodex
Spec pipeline conductor — run the Spec Kit chain (specify→clarify→plan→tasks→analyze) with blocking gates, enforce a verify-criteria task contract, and hand off a commit-ready plan. Use when turning a feature idea into an executable, test-gated plan.
Skill Claude CodeCodex
Smart pull requests — Conventional Commit title and clean body for release-please. Use when opening or formatting a pull request.
Skill Claude CodeCodex
Paid-product readiness — monetization model, billing/entitlement integrity, pricing/packaging, go-to-market baseline. Use when turning a project into a paid product or auditing an existing monetization setup.
Skill Claude CodeCodex
Installs a deterministic, local, no-CI quality gate (format/lint/type/test) wired into a host-appropriate hook, plus --invariant mode turning any described invariant into a red-proven, chain-wired check. Use to enforce quality, block "done" without CI, or mechanize an invariant.
Skill Claude CodeCodex
Repository health — audit and configure repo settings, branch policies, hygiene, metadata, team, and structure. Use when setting up or auditing repository configuration.
Skill Claude CodeCodex
Smart research — parallel search, tier sources, synthesize, and recommend. Use when the user wants quick multi-source research and a recommendation.
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: