revskills

Checkpoint checklist for RevFleet sessions. Validates the 6 coherent-tracking surfaces, inventories tracking state, writes a rolling handoff fragment + workboard log fragment, re-renders CURRENT-HANDOFF/workboard locally for read convenience, and commits ONLY append-only fragments (docs/handoffs/rolling + .claude/workboard.d) per ADR 2026-07-23-jv-coordination-merge-model. Worktree-gated when a peer is live. Never commits derived CURRENT-HANDOFF.md or workboard.md. Never master-handoff regen or auto-merge with --admin.

Multi-session exhaustive codebase audit framework. Inventory every file path with hashes/line counts, partition into claimable shards for concurrent or serial agents, read and verify every line with a machine-checked coverage ledger, cross-check docs/config/code, and optionally map the system via the knowledge graph (revkg / kg_*). Use when the user runs /exhaustive-audit, asks for a full-codebase audit, line-level verification, multi-agent audit, fleet-wide audit, or "account for every file/line".

Fleet workflow for finding duplication, deprecation, and redundancy (exact file clones, deprecation markers, lockstep/mirror comments, basename collisions). Multi-session friendly with machine findings + human classification (intentional vs accidental). Use when /redundancy-scan, /dedupe, fleet redundancy audit, find duplicates, dead dual implementations, deprecated code still shipping, or extend-before-create consolidation passes.

Query the fleet knowledge graph before grep for cross-repo questions (what depends on X, where is Y enforced, what did the dependency graph look like on date D, what has any agent learned about Z). Covers the kg_* MCP tool surface, kg_context as the default entry point, point-in-time queries, publishing agent-fact episodes, and the revkg CLI equivalents. GAP-349.

A minimal valid skill fixture used to prove the skills-lint and plugin-lint validators accept well formed input. Covers frontmatter shape only.

Verify the .claude/rules distribution topology across the RevealUI repos (symlinks into revcon profiles, or materialized copies gated in-repo). Read-only; never mutates, always asks before any fix.

Capture a mid-session fidelity snapshot for RevFleet sessions, keyed to this session's id. Assembles mechanical state (branch, dirty, ahead, open PRs) and has the agent author the five sections the checkpoint skill consumes (Resume-From-Here, What-Shipped, Active-Constraints, Do-Not-Repeat, Open-Loose-Ends), writing ~/.claude/coordination/snapshots/$CLAUDE_CODE_SESSION_ID.md. Also promotes durable feedback/rule-class lessons to the memory directory. Exposed as /snapshot. Nudged by the track-session context advisory at the soft-context line; the producer half of GAP-317.

Thin shim onto the operational-workflow-layer runner (workflow-run.js). Lists registered workflows (cleanup-session, prepare-for-exit) or runs one by name, passing --dry-run/--fix/--yes straight through. Exposed as /ops. Delegates entirely to the runner — never reimplements workflow execution or safety classification (auto/report-first/gated/owner-only).

Static validator for Claude Code skills. Dry-runs every skill in ~/.claude/commands/*.md, validates referenced scripts exist, flags stale ~/suite/ paths, rule violations (git -C, pnpm --dir, inline node -e), tmux references. Read-only — catches broken skills before users hit them at run time.

Check whether the claude.ai/design project changed since the codebase last pushed to it, and which files. Design-to-code awareness half of the design-code sync loop (GAP-322). Compares the committed sync-state file against DesignSync list_projects / list_files / get_file, diffs against the design project's own _sync_manifest.json, and reports new/deleted/modified files as a proposal-shaped next step (never writes code or pushes to design from this skill).

Use this skill to generate well-branded interfaces and assets for RevealUI Studio, either for production or throwaway prototypes/mocks/etc. Contains essential design guidelines, colors, type, and brand voice rules for the RevealUI design system (Cobalt brand, dark-first OKLCH tokens). Read README.md for the full canonical doc and CLAUDE.md for brand orientation, voice rules, and assessment summaries.

Drizzle ORM schema design, migrations, and query patterns for PostgreSQL. Use when creating tables, writing queries, defining relations, managing migrations, or working with NeonDB. Covers pgTable, indexes, enums, jsonb columns, transactions, soft delete, pagination, drizzle-kit generate/migrate, and serverless connection patterns.

ElectricSQL v1.x real-time sync patterns for PostgreSQL. Use when setting up shape subscriptions, authenticated proxy routes, mutation endpoints, offline-first sync, or connecting Electric to Neon. Covers read-only sync, row-level filtering, write-through-API, and conflict resolution.

Health check for RevFleet Claude Code setup + Studio-native workflow. Verifies hook syntax, rules directories, skill preconditions, git integrity across RevFleet repos, workboard freshness, daemon status, MCP servers, env file leaks, settings JSON validity, toolchain, and disaster-recovery snapshot state.

DEPRECATED 2026-07-02 — the per-repo LTS sync model (bundle/mirror declared in .claude/lts-mode) is retired. Disaster recovery is the weekly whole-distro WSL snapshot (Windows scheduled task RevealUI-WSL-Weekly-Backup, shipped as revkit scripts/weekly-wsl-backup.ps1 with staleness alerting). Retained for reference; will be removed in a future revskills major.

Diagnose and recover from a crashed or interrupted Claude/Studio session in RevFleet. Surfaces git corruption, stale hook state, orphaned handoffs, daemon status, and workboard CRASHED markers. Diagnostic-first — never executes destructive repairs without explicit approval.

Model Context Protocol server development patterns. Use when building custom MCP servers, exposing tools and resources, handling JSON-RPC, implementing credential isolation, or integrating with Claude Desktop, Cursor, or AI agents. Covers @modelcontextprotocol/sdk, StdioServerTransport, tool schemas, and multi-tenant credential patterns.

Multi-agent shared memory patterns for AI agent coordination. Use when building shared fact logs, collaborative scratchpads, memory reconciliation, or coordinating discoveries between concurrent agents. Covers append-only fact tables, Yjs CRDT scratchpads, LLM-powered reconciliation, and session-scoped memory sharing.

Yjs CRDT collaboration patterns for real-time and offline-first editing. Use when building collaborative documents, shared scratchpads, or multi-user editing. Covers Y.Doc, Y.Map, Y.Array, Y.Text, state encoding, structured patches, server-side document manipulation, and conflict-free merge semantics.

Next.js 15+ App Router best practices for pages, layouts, routes, server components, server actions, caching, API routes, and streaming. Use when building or reviewing Next.js code, implementing RSC patterns, PPR, metadata, error boundaries, or proxy.ts. Covers React 19 conventions, data fetching, ISR, and common mistakes.

Web application security hardening for OWASP Top 10 vulnerabilities. Use when implementing authentication, HTTP security headers, CORS, rate limiting, input validation, or reviewing code for XSS, SQL injection, CSRF, or SSRF. Covers CSP, HSTS, session cookies, bcrypt, Zod validation, and secrets management.

Tailwind CSS v4 patterns, CSS-first configuration, and v3 migration guide. Use when writing styles, configuring @theme variables, creating components with CVA, migrating from tailwind.config.js, or using container queries, dark mode, and new v4 utility syntax. Covers @import tailwindcss, custom properties, oklch colors, and clsx.

Vitest testing patterns for production TypeScript projects. Use when writing tests, fixing test failures, configuring coverage, mocking modules, debugging flaky tests, or setting up Vitest in a monorepo. Covers Vitest 3+, vi.mock, vi.fn, React Testing Library, PGlite in-memory PostgreSQL, pool forks, maxWorkers, and hookTimeout.