Help us improve
Share bugs, ideas, or general feedback.
npx claudepluginhub forbee-dev/forgebeeBatteries-included development and growth framework for Claude Code — 33 slash commands, 69 specialist agents (including 12 review sub-agents), 26 lifecycle hooks across 9 events, 3 skills, mode-aware permission system (auto/bypass/default with non-negotiable blocklist), adversarial debate system with batched debates (dev + strategy), continuous learning with heuristic pattern detection and pending-instinct approval workflow, command-to-agent delegation with fallback, manifest-driven sync tooling, JSON handoff contracts, anti-rationalization gates, evidence-based verification gates, TDD enforcement, crash recovery checkpointing, governance audit trail, agent contract validation, Growth OS 9-phase marketing pipeline with CRO and email automation, and automated project management.
Claude Code marketplace entries for the plugin-safe Antigravity Awesome Skills library and its compatible editorial bundles.
Production-ready workflow orchestration with 84 marketplace plugins, 192 local specialized agents, and 156 local skills - optimized for granular installation and minimal token usage
Directory of popular Claude Code extensions including development tools, productivity plugins, and MCP integrations
Share bugs, ideas, or general feedback.
A colony of AI agents forging your product
69 specialist agents. 33 slash commands. 26 lifecycle hooks. 3 skills.
Mode-aware permissions. Continuous learning. Adversarial debate. Growth OS.
Command-to-agent delegation. Manifest-driven sync. Pure Node.js. Zero deps.
Works with Claude Code and OpenClaw.
Quick Start • Commands • Agents • Hooks • Learning • Growth OS • PM System • OpenClaw
Claude Code and OpenClaw are powerful out of the box. ForgeBee makes them opinionated and structured — so your agent plans before it codes, debates before it ships, and verifies before it marks anything "done."
| Without ForgeBee | With ForgeBee |
|---|---|
| Agent jumps straight into coding | Agent plans, debates requirements, then codes |
| "It should work" | Evidence-based verification with actual test output |
| Single-agent, single-pass | 69 specialists working in parallel with blind review |
| Manual project tracking | Automated state.yaml + markdown dashboards |
| No marketing workflow | Full 9-phase Growth OS with 13 marketing agents + 3 strategy debate agents |
| Every session starts from scratch | Continuous learning — heuristic pattern detection + pending instinct approval |
| Edits break silently | Auto-format, typecheck, and lint on every edit |
| Permission prompts everywhere | Mode-aware permissions — respects auto-mode and bypass with non-negotiable blocklist |
| Commands duplicate agent logic | Commands delegate to specialist agents with automatic fallback |
| Review finds issues that should have been caught | Quality pipeline — specialists self-review, code-skeptic validates, review-all just confirms |
| Agents take shortcuts under pressure | Every agent has explicit "Never" rules — hard boundaries that can't be rationalized away |
Two commands. That's it.
# Inside Claude Code:
/plugin marketplace add forbee-dev/ForgeBee
/plugin install forgebee@forbee-dev
Then edit CLAUDE.md with your project details and start working. Hooks activate automatically, commands are available immediately.
git clone git@github.com:forbee-dev/ForgeBee.git
claude --plugin-dir ./ForgeBee/forgebee
CLAUDE_CODE_EXPERIMENTAL_AGENT_TEAMS=1Invoke with a slash: /review, /debug, /workflow, etc.
| Command | Description |
|---|---|
/plan | Phased planning: Brief → Requirements → Architecture → Sprint Stories |