Skip to content

ForgeDock Documentation

ForgeDock is an autonomous development pipeline for Claude Code. You open a GitHub issue, type /work-on, and an AI agent investigates the problem, architects a solution, builds it, runs quality checks, reviews it with 9 specialized agents, and merges the PR.

No more coordinating between agent sessions. No more lost context. No more starting from scratch.


The core idea

GitHub Is Already Your Agents' Memory — Why agents re-derive everything from scratch every session, why sidecar stores duplicate what GitHub already records, and how FORGE annotations make the existing citation graph machine-readable. This is the canonical explanation of why ForgeDock works the way it does.


Get Started

New to ForgeDock? Start here:


Guides

PageWhat You'll Learn
GitHub Is Already Your Agents' MemoryThe memory problem, FORGE annotations, the knowledge graph argument, and the open protocol
Getting StartedInstall ForgeDock, configure forge.yaml, run /work-on on your first issue
How It WorksFORGE annotations, the pipeline relay, compaction resilience, workflow labels
ForgeDock vs. Manual WorkflowsWhy structured pipelines beat ad-hoc Claude Code sessions
FORGE Annotation ProtocolTechnical spec — annotation types, completion markers, querying
Command ReferenceAll 25+ commands with usage, options, and examples
For CompaniesAGPL vs. commercial license, fleet layer, design-partner program, procurement facts


Install

bash
npx forgedock

Installs commands into ~/.claude/commands/, available in every Claude Code session on this machine. --global is still accepted as a flag for backward compatibility, but install is always global — there's nothing to opt into.

Requirements: Claude Code, Node.js 18+, gh CLI authenticated with GitHub.


License

ForgeDock is open source under the AGPL-3.0 license.

Released under the AGPL-3.0 License.