By ascend42
AI-native project architecture — structured context, enforcement hooks, multi-agent orchestration, and 50+ MCP tools for symbol-driven development
System design and architecture planning agent. Use when tasks require architectural decisions, designing multi-file features, choosing between implementation approaches, or planning changes that span 3+ components. Read-only — does not write implementation code.
Implementation and coding agent. Use for writing code, creating files, and making changes based on specifications or task descriptions. Follows existing patterns, updates .purpose files alongside code, and uses the Paradigm logger instead of console.log.
Code review and quality analysis agent. Use to review changes for bugs, spec compliance, code smells, and convention adherence. Read-only — does not modify code. Uses a two-stage protocol: spec compliance first, then code quality.
Security analysis agent for authorization, authentication, input validation, and vulnerability detection. Use when tasks involve auth, user data, API endpoints, or when reviewing code for security issues. Read-only — flags issues for Builder to fix.
Testing and validation agent. Use after implementation to write tests, run test suites, verify edge cases, check gate validations, and ensure code quality. Can read and write test files, and run test commands.
Manage your agent team — roster, onboard, bench/activate, detail, install from GitHub or nevr.land. Use when the user says "agents", "show my agents", "onboard agents", "install agent", "bench agent", or wants to manage their AI team.
Register this Claude Code session with Paradigm Conductor. Use when the user says "conduct", "surface to conductor", "register with conductor", or wants this terminal visible in the Conductor overlay.
Run a comprehensive health check on the Paradigm setup. Use when something seems wrong, after major changes, or when the user says "check paradigm health", "paradigm doctor", or "is paradigm set up correctly".
Prepare a context handoff summary for the next session or agent. Use when the user says "prepare handoff", "I'm done for now", "hand this off", or when context usage is high and work needs to continue later.
Agent learning health dashboard — Neverland metrics, nomination stats, threshold drift, notebook growth. Use when the user says "health", "agent health", "neverland", "learning metrics", "how are agents doing", or wants to check the learning system.
Executes bash commands
Hook triggers when Bash tool is used
Modifies files
Hook triggers on file write and edit operations
Own this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimOwn this plugin?
Verify ownership to unlock analytics, metadata editing, and a verified badge. GitHub access is read-only (username + org membership).
Sign in to claimBased on adoption, maintenance, documentation, and repository signals. Not a security audit or endorsement.
Uses power tools
Uses Bash, Write, or Edit tools
Uses power tools
Uses Bash, Write, or Edit tools
One command to make your codebase AI-ready.
npm install -g @a-company/paradigm
cd your-project
paradigm shift
That's it. paradigm shift scaffolds everything your AI assistant needs to understand your codebase — a project map, access rules, IDE instruction files, and enforcement hooks — in one command. Safe to re-run. Works with TypeScript, Python, Rust, Go, Swift, and more.
┌─────────────────────────────────────────────────┐
│ paradigm shift │
│ Full project setup in one command │
└─────────────────────────────────────────────────┘
📁 Project: your-project
📍 Status: New project
✓ Step 1/6: Initializing Paradigm...
✓ Step 2/6: Initializing team configuration...
✓ Step 3/6: Scanning and indexing symbols...
✓ Step 4/6: Syncing IDE configurations...
✓ Step 5/6: Installing hooks...
✓ Step 6/6: Skipped verify (use --verify to check health)
┌─────────────────────────────────────────────────┐
│ ✨ Paradigm shift complete! │
└─────────────────────────────────────────────────┘
Created/Updated:
─────────────────────────────────────────────────
✓ .paradigm/config.yaml Project configuration
✓ .paradigm/navigator.yaml Symbol navigation map
✓ .paradigm/agents.yaml Team agent configuration
✓ .purpose Root feature definitions
✓ portal.yaml Authorization gates
✓ CLAUDE.md Claude Code AI instructions
✓ AGENTS.md Universal AI agent instructions
✓ .cursor/rules/ Cursor AI instructions
✓ .claude/hooks/ Claude Code enforcement hooks
Skip to what you just got → · Skip to the why →
curl -fsSL https://a-company.org/paradigm/install.sh | bash
Or download and inspect first:
curl -fsSL https://a-company.org/paradigm/install.sh -o install.sh
chmod +x install.sh
./install.sh
This clones the repo to ~/.paradigm-cli/, builds both paradigm and paradigm-mcp, and installs them globally. Re-running updates to the latest version.
Note: The source directory at
~/.paradigm-cli/must be kept — the global CLIs symlink to it. To uninstall:npm uninstall -g @a-company/paradigm @a-company/paradigm-mcp && rm -rf ~/.paradigm-cli
# 1. Clone and build (keep this directory — CLIs symlink to it)
git clone https://github.com/ascend42/a-paradigm.git ~/.paradigm-cli
cd ~/.paradigm-cli
npm install && npm run build
# 2. Install CLI globally
cd packages/paradigm && npm install -g . && cd ../..
# 3. Install MCP server globally
cd packages/paradigm-mcp && npm install -g . && cd ../..
# 4. Verify
paradigm --version
If you use Claude Code, the plugin bundles enforcement hooks and MCP configuration in a single install:
/plugin marketplace add ascend42/a-paradigm
You still run paradigm shift inside each project to create the per-project files (.paradigm/, .purpose, portal.yaml, etc.) — the plugin handles the global pieces so you don't need to run paradigm mcp setup separately.
paradigm shift — the one commandRun it on a fresh project or an existing one. It detects which case it's in and does the right thing. In plain English:
.paradigm/ with project config, agent team, and IDE detection.purpose (what your code does) and portal.yaml (who can access what) as empty skeletons you fill in--quick)CLAUDE.md, AGENTS.md, .cursor/rules/, Windsurf, Copilot)paradigm doctor to verify everything is healthy (--verify)Re-running is safe. 36 core files and directories are guaranteed to exist after completion, and existing content is never overwritten.
npx claudepluginhub ascend42/a-paradigm --plugin paradigmHarness-native ECC operator layer - 67 agents, 271 skills, 92 legacy command shims, reusable hooks, rules, selective install profiles, and production-ready workflows for Claude Code, Codex, OpenCode, Cursor, and related agent harnesses
Reliable automation, in-depth debugging, and performance analysis in Chrome using Chrome DevTools and Puppeteer
Feature development with code-architect/explorer/reviewer agents, CLAUDE.md audit and session learnings, and Agent Skills creation with eval benchmarking from Anthropic.
Access thousands of AI prompts and skills directly in your AI coding assistant. Search prompts, discover skills, save your own, and improve prompts with AI.
Intelligent draw.io diagramming plugin with AI-powered diagram generation, multi-platform embedding (GitHub, Confluence, Azure DevOps, Notion, Teams, Harness), conditional formatting, live data binding, and MCP server integration for programmatic diagram creation and management.
Comprehensive feature development workflow with specialized agents for codebase exploration, architecture design, and quality review