Forge-Core
AGENTS.md2 starsChanged 5 days agoMITView on GitHub
The agents are Claude Code subagents defined in .claude/agents/*.md. Each file sets the agent's role, the tools it may use, and its default model and effort. /forge picks the smallest set that fits the task; a one-line fix may use a single agent. The plugin includes these agents as well. Forge can also hand work to other Claude Code agents you have installed yourself; those are not part of this repository.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- ForgeyClap/claude-forgeword for word, first seen Nov 1, 2026
Other versions1 project uses a version of this.
Word for word · 1
- ForgeyClap/claude-forge★ 2100% same
Edited · 0
None found.
Wrote their own version · 0
None found.
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository99 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- brainstormingSkill
- claude-md-improverSkill
- dispatching-parallel-agentsSkill
- executing-plansSkill
- finishing-a-development-branchSkill
- forge-a11ySkill
- forge-agent-reportSkill
- forge-agentSkill
- forge-apiSkill
- forge-authSkill
- forge-autoresearchSkill
- forge-botsSkill
- forge-brainstormSkill
- forge-cliSkill
- forge-cmsSkill
- forge-code-reviewSkill
- forge-contextSkill
- forge-councilSkill
- forge-craftSkill
- forge-databaseSkill
- forge-dataSkill
- forge-debugSkill
- forge-deeplearnSkill
- forge-devopsSkill
- forge-docdriftSkill
- forge-doctorSkill
- forge-ecc-adapterSkill
- forge-ecommerceSkill
- forge-electronSkill
- forge-extensionSkill
- forge-figmaSkill
- forge-fullstackSkill
- forge-gameSkill
- forge-graded-verifySkill
- forge-guardianSkill
- forge-heartbeatSkill
- forge-i18n-docsSkill
- forge-intakeSkill
- forge-integrationSkill
- forge-long-form-writingSkill
- forge-mcp-clientsSkill
- forge-memorySkill
- forge-migrationSkill
- forge-mindmapSkill
- forge-mlopsSkill
- forge-mobileSkill
- forge-n8nSkill
- forge-nightshiftSkill
- forge-observabilitySkill
- forge-paymentsSkill
- forge-plan-largeSkill
- forge-prdSkill
- forge-predictionSkill
- forge-projectbrainSkill
- forge-prompt-coachSkill
- forge-qualitySkill
- forge-ragSkill
- forge-registrySkill
- forge-releaseSkill
- The first 60 shown.
The file
Files
Agents
The agents are Claude Code subagents defined in .claude/agents/*.md. Each file sets the agent's role, the tools it may use, and its default model and effort. /forge picks the smallest set that fits the task; a one-line fix may use a single agent.
How a task moves through them
bossowns a larger task and splits it into goals.head-chefturns those goals into concrete work packages, gives each a route (how much model and effort it needs) and checks that each package was actually done.- The specialists below do the work.
test-bossruns the real tests.verify-bossre-runs the commands a worker says passed and replies PASS or FAIL.review-bosscompares the result with what you asked for before Forge reports back. A failed check gets one more attempt with more effort or a stronger model; a second failure is reported to you.
The agents
| File | Role |
boss.md | Plans a larger task and decides how to handle a failed check. |
head-chef.md | Splits the plan into work packages and checks that each one was completed. |
build-boss.md | Writes and changes code. |
ui-boss.md | Front-end and layout work, checked with screenshots at phone, tablet and desktop widths. |
integration-boss.md | APIs, webhooks, n8n and other external services, with retries, timeouts and clear errors. |
test-boss.md | Runs the automated tests (Playwright for web apps, the matching runner elsewhere). |
verify-boss.md | Re-runs the build, test and health-check commands another agent reported, read and execute only. |
review-boss.md | Final review against the original request: code, logic, tests, UX and missing pieces. |
security-boss.md | Read-only review of keys, environment variables, authentication, webhooks, input handling and logging. |
seo-boss.md | Website metadata, structure, performance, accessibility and indexability. |
search-boss.md | Research with sources, for questions that need current information. |
docs-boss.md | Setup instructions, architecture notes and the final report. |
skill-boss.md | Keeps the skill registry and assigns skills to agents. |
scribe.md | Records each instruction you give verbatim and tracks whether it was done. |
codex-reviewer.md | Optional second review through the OpenAI Codex CLI, when it is installed. |
data-scientist.md | Data analysis and statistical models, with stated uncertainty. |
ml-engineer.md | Training-to-serving pipelines, model versioning and monitoring. |
electron-pro.md | Electron desktop apps: safe IPC, context isolation and a working installer. |
mcp-developer.md | Model Context Protocol servers and clients. |
payment-integration.md | Payment providers, verified webhooks and idempotent charges and refunds. |
The plugin includes these agents as well. Forge can also hand work to other Claude Code agents you have installed yourself; those are not part of this repository.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.