{"id":51548806,"url":"https://github.com/samzong/agent-brains","last_synced_at":"2026-07-09T21:30:37.096Z","repository":{"id":363474453,"uuid":"1263010338","full_name":"samzong/agent-brains","owner":"samzong","description":"Personal agent brain catalog for coding agents: AGENTS.md, SOUL.md, MEMORY.md, skills, workflows, and a bundled loader.","archived":false,"fork":false,"pushed_at":"2026-06-09T03:59:36.000Z","size":50,"stargazers_count":1,"open_issues_count":0,"forks_count":0,"subscribers_count":0,"default_branch":"main","last_synced_at":"2026-06-09T05:09:27.149Z","etag":null,"topics":["agent-brains","agent-skills","agents-md","ai-agents","claude-code","codex","coding-agents","cursor"],"latest_commit_sha":null,"homepage":null,"language":"Python","has_issues":true,"has_wiki":null,"has_pages":null,"mirror_url":null,"source_name":null,"license":"mit","status":null,"scm":"git","pull_requests_enabled":true,"icon_url":"https://github.com/samzong.png","metadata":{"files":{"readme":"README.md","changelog":null,"contributing":null,"funding":null,"license":"LICENSE","code_of_conduct":null,"threat_model":null,"audit":null,"citation":null,"codeowners":null,"security":null,"support":null,"governance":null,"roadmap":null,"authors":null,"dei":null,"publiccode":null,"codemeta":null,"zenodo":null,"notice":null,"maintainers":null,"copyright":null,"agents":"AGENTS.md","dco":null,"cla":null}},"created_at":"2026-06-08T14:33:24.000Z","updated_at":"2026-06-09T03:59:40.000Z","dependencies_parsed_at":null,"dependency_job_id":null,"html_url":"https://github.com/samzong/agent-brains","commit_stats":null,"previous_names":["samzong/agent-brains"],"tags_count":null,"template":false,"template_full_name":null,"purl":"pkg:github/samzong/agent-brains","repository_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/samzong%2Fagent-brains","tags_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/samzong%2Fagent-brains/tags","releases_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/samzong%2Fagent-brains/releases","manifests_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/samzong%2Fagent-brains/manifests","owner_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners/samzong","download_url":"https://codeload.github.com/samzong/agent-brains/tar.gz/refs/heads/main","sbom_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories/samzong%2Fagent-brains/sbom","scorecard":null,"host":{"name":"GitHub","url":"https://github.com","kind":"github","repositories_count":286080680,"owners_count":35313755,"icon_url":"https://github.com/github.png","version":null,"created_at":"2022-05-30T11:31:42.601Z","updated_at":"2026-05-26T15:22:16.424Z","status":"online","status_checked_at":"2026-07-09T02:00:07.329Z","response_time":57,"last_error":null,"robots_txt_status":"success","robots_txt_updated_at":"2025-07-24T06:49:26.215Z","robots_txt_url":"https://github.com/robots.txt","online":true,"can_crawl_api":true,"host_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub","repositories_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repositories","repository_names_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/repository_names","owners_url":"https://repos.ecosyste.ms/api/v1/hosts/GitHub/owners"}},"keywords":["agent-brains","agent-skills","agents-md","ai-agents","claude-code","codex","coding-agents","cursor"],"created_at":"2026-07-09T21:30:36.412Z","updated_at":"2026-07-09T21:30:37.077Z","avatar_url":"https://github.com/samzong.png","language":"Python","funding_links":[],"categories":[],"sub_categories":[],"readme":"# agent-brains\n\nPortable **domain agent brains** for coding agents (Codex, Claude Code, Cursor, and others that read markdown context).\n\nAn agent brain is a long-lived context pack for a recurring domain. It keeps identity, judgment criteria, style, durable memory, domain methods, and stable SOPs in one loadable structure so the agent does not start from a blank context every session.\n\nSee [docs/AGENT_BRAIN_SPEC.md](./docs/AGENT_BRAIN_SPEC.md) for the full specification.\n\n**Skill vs workflow vs brain** (see spec for detail):\n\n- `skill` — reusable method or tool capability (**how**)\n- `workflow` — end-to-end SOP for stable business behavior (**what process**)\n- `agent brain` — identity, memory, and domain judgment (**who**, on **what history and standards**)\n\n## Create brains\n\n| Name | Purpose |\n|------|---------|\n| `agent-creator` | Create, audit, and evolve agent brains |\n\n## Bundled skills\n\n| Name | Purpose |\n|------|---------|\n| `agent-loader` | List and load repo-scoped and user-scoped agent brains |\n\nBuilt-in loader: [`skills/agent-loader/`](./skills/agent-loader/). Copy or symlink it into your agent client's global skills directory to use it outside this repo.\n\nDiscovery checks repo-local `.agents/agents` directories from the current working directory up to the repository root, then `$HOME/.agents/agents`. It reads only `AGENTS.md` frontmatter and reports `name`, `description`, `scope`, and `path`. On load, `agent-loader` resolves the agent path, then reads `AGENTS.md`, `SOUL.md`, and `MEMORY.md` first. `memories/`, `skills/`, and `workflows/` are lazy-loaded only when the task needs them or core files point to them.\n\n## Optional local routing\n\nFor Codex users, [`codex-agents-local`](https://github.com/samzong/codex-agents-local) can add a private `AGENTS.local.md` prompt that nudges Codex to choose an agent brain only when useful:\n\n```md\nWhen a task is domain-specific or recurring enough to benefit from a personal agent brain, proactively use `agent-loader` to list/resolve candidates, load the single best-matching agent core before working, ask only if the match is ambiguous, and keep skills/workflows/memories lazy-loaded.\n```\n\n## Brain structure\n\nEach brain lives under `agents/`:\n\n```text\nagents/\n└── agent-name/\n    ├── AGENTS.md\n    ├── SOUL.md\n    ├── MEMORY.md\n    ├── memories/\n    ├── skills/\n    │   └── SKILL_NAME/\n    │       └── SKILL.md\n    └── workflows/\n        └── WORKFLOW_NAME/\n            └── WORKFLOW.md\n```\n\n- `AGENTS.md` — identity, responsibilities, rules, boundaries (`name`, `description` frontmatter required)\n- `SOUL.md` — style, judgment, taste, and tone\n- `MEMORY.md` — durable main memory\n- `memories/` — optional focused sub-memories (by date or topic)\n- `skills/\u003cskill-name\u003e/SKILL.md` — optional reusable methods and tools\n- `workflows/\u003cworkflow-name\u003e/WORKFLOW.md` — optional end-to-end SOPs\n\nKeep brains narrow. ***A useful brain knows one domain well instead of trying to manage every task.***\n\n## Related projects\n\n- [AGENTS.md](https://agents.md) — open format for agent-facing project instructions\n- [OpenClaw workspace](https://docs.openclaw.ai/concepts/agent) — runtime-injected workspace files (similar file semantics, different loading model)\n\n## License\n\nMIT\n","project_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsamzong%2Fagent-brains","html_url":"https://awesome.ecosyste.ms/projects/github.com%2Fsamzong%2Fagent-brains","lists_url":"https://awesome.ecosyste.ms/api/v1/projects/github.com%2Fsamzong%2Fagent-brains/lists"}