The Composure Suite
One install. Six systems. Zero configuration.
6 plugins. 37 skills. Each specialized — together, they share one code graph, one task queue, and one commit gate.
Composure
Architecture & code quality
17 skillsView plugin
Skills
/composure:initialize— One-time project bootstrap. Detects tech stack, queries Context7 for framework docs, builds code review graph, creates task queue infrastructure./composure:build-graph— Creates or updates the code knowledge graph. Parses source files with tree-sitter, maps imports, identifies functions/types/components, builds dependency graph./composure:blueprint— Graph-powered planning with progressive refinement. Classifies work, scans code graph for related files, confirms scope with user, analyzes impact, writes implementation spec./composure:review-pr— Comprehensive PR review using the code graph. Identifies changes, resolves linked issues, builds structural context, generates walkthrough, runs quality audit./composure:review— Token-efficient review of only changed code and its blast radius. Uses graph to auto-detect impact./composure:audit— Full audit across 5 categories: Architecture, Security, Code Quality, Dependencies, Test Coverage. Letter grades A-F./composure:code-organizer— Restructures disorganized projects into conventional layouts. Detects file types, maps to appropriate directories, executes moves with import updates./composure:report— Orchestrates all plugins to generate a self-contained HTML audit report with letter grades, tabbed sections, and prioritized recommendations./composure:commit— Commits with automatic task queue hygiene. Cleans resolved tasks, archives completed audits, blocks if staged files have open Critical/High items./composure:app-architecture— Master router to framework-specific architecture guides. Detects stack, routes to frontend/fullstack/mobile/backend docs./composure:backlog— Processes accumulated code quality tasks. Modes: batch (sequential), delegate (parallel sub-agents), verify, archive, clean, sync./composure:changelog— Auto-generates changelog from git history enriched with code graph entity mapping. Groups by conventional commit type./composure:view-graph— Opens interactive HTML visualization in browser. Files grouped by category, entity relationships, impact radius from selected nodes./composure:update-project— Lightweight refresh for already-initialized projects. Updates version stamp, re-detects stack, regenerates stale docs./composure:integration-builder— Stack-agnostic third-party integration scaffolding. Detects your project language, pulls latest SDK docs via Context7, and scaffolds client, auth, webhooks with proper error handling./composure:mcp-builder— Build custom MCP servers to connect any service to Claude Code. Discovers the target API, designs tool definitions, and scaffolds a complete TypeScript MCP server project./composure:mcp-setup— Discover and configure existing MCP servers for your project. Stack-aware recommendations from a curated registry with full setup — install, configure, and verify.Sentinel
Security scanning
5 skillsView plugin
Skills
/sentinel:assess— Assesses your project's security surface — detects stack, package managers, security tooling, and integrations. Generates config for ongoing scanning./sentinel:scan— Combines Semgrep static analysis and dependency audit. Writes findings to tasks-plans/tasks.md with severity mapping./sentinel:audit-deps— Focused vulnerability audit. Reports CVEs with installed/fixed versions and exact upgrade commands./sentinel:headers— Analyzes security headers for a given URL. Context-aware grading based on exploitable risk, not checkbox compliance./sentinel:package-risk— Analyzes an installed package's source code for suspicious behavior patterns — eval calls, network requests, environment access, and obfuscation.Shipyard
CI/CD & deployment
6 skillsView plugin
Skills
/shipyard:configure— Detects CI/CD platforms, deployment targets, container configs. Generates .claude/shipyard.json./shipyard:ci-generate— Generates production-ready CI/CD workflows from detected stack. Shows diff before overwriting existing workflows./shipyard:ci-validate— Validates workflows using actionlint + built-in heuristic checks./shipyard:dockerfile— Generates or validates Dockerfiles with security best practices. Framework-specific multi-stage builds./shipyard:deps-check— Broader than Sentinel — includes outdated packages. Determines highest-safe-version accounting for cascading vulnerabilities./shipyard:preflight— Production readiness checklist. Checks environment variables, health endpoints, error monitoring, CORS, rate limiting, and security headers.Testbench
Convention-aware testing
3 skillsView plugin
Skills
/testbench:calibrate— Detects test framework (Vitest, Jest, Playwright, Cypress, pytest, Go test, Cargo test, RSpec, PHPUnit). Reads conventions. Generates config./testbench:generate— Generates tests for a single file. Reads config AND existing test files, never generates without context./testbench:run— Runs tests with scope control: all, changed files only, or single-file. Parses output, extracts failures with source context.Design Forge
Premium design patterns
2 skillsView plugin
Skills
/design-forge:design-forge— Premium design patterns and production-ready components. Canvas visualizations, glassmorphism, animations, interactive experiences./design-forge:ux-researcher— AI-powered design research agent using WebSearch and WebFetch to discover patterns, evaluate competitors, analyze technologies.Cortex
AI memory & reasoning persistence
4 skillsView plugin
Skills
/composure:cortex— Direct interface to the Cortex memory system. Search, create, traverse, and manage memory nodes and edges.sequential_think— Persistent chain-of-thought reasoning that survives across sessions. Pick up complex analysis exactly where you left off — even weeks later.search_memory— A knowledge network that grows with every session. Nodes and edges form a project-aware memory graph linking decisions, patterns, and learnings.capture_thought— Automatic behaviors that require no commands. Cortex captures reasoning when you use sequential thinking, nudges task creation on complex prompts, and loads relevant memories at session start.