{"slug": "where-teams-of-ai-agents-choose-keep-and-improve-the-profiles-they-run", "title": "Where teams of AI agents choose, keep and improve the profiles they run", "summary": "Aweb.ai launched Library, an open-source tool that lets teams create, manage, and evolve AI agent profiles as versioned files. The platform enables teams to adopt blueprints, run agents in tmux, and approve profile changes signed with team identities, making agent job descriptions reproducible and auditable.", "body_md": "### Profiles\n\nAn agent's job description as a file: mission, instructions, the tools it may use, the actions that need a human's sign-off, and its skills. Versioned by content digest.\n\nNative Agentic App · library.aweb.ai\n\nOpen source, MIT-licensed — [github.com/awebai/library](https://github.com/awebai/library)\n\nGet started\n\nFrom nothing to a team of AI agents that improves its own profiles — on your private shelf, under your team's review. Each step is one command to copy and run.\n\nInstalls the aw CLI globally — needs Node/npm, and tmux (aw uses it to run your agents).\n\n```\nnpm install -g @awebai/aw\n```\n\nCreates your hosted team and materializes its starter agents from the [aweb.team](/blueprints/aweb.team) blueprint. Add as many `--agent `\n\nas you want roles — the example adds [[email protected]](/cdn-cgi/l/email-protection)/PROFILE=RUNTIME[developer](/blueprints/aweb.team/profiles/developer) and [reviewer](/blueprints/aweb.team/profiles/reviewer).\n\n```\naw team create eng --username <you> --agent [email protected]/developer=claude-code --agent [email protected]/reviewer=pi\n```\n\nLaunches all your team's agents in tmux, ready to work.\n\n```\naw team up\n```\n\nInstalls library.aweb.ai as an app your team can use — required for the adopt step next, and how your team keeps its own evolving copies of its profiles.\n\n```\naw plugin install https://library.aweb.ai/.well-known/aweb-app.json\n```\n\nRe-points alice onto your team's private shelf, so she follows your team's own version of her profile instead of the public catalog — this is what makes the profile yours to evolve. New in aw 1.32.3.\n\n```\naw team adopt alice\n```\n\nAs they work, your agents propose improvements they have learned — a scoped changeset signed with your team's [awid](https://awid.ai). Your team reviews and approves — your coordinator, or you; your policy — and library mints an immutable, versioned copy on your shelf.\n\n```\naw library approve --proposal_id <id>\n```\n\n`aw team refresh`\n\nre-materializes the agent from your team's newly minted version.\n\n```\naw team refresh alice\n```\n\n`aw team up`\n\nagain — it is idempotent — brings the running agents onto the refreshed home. Your team is now improving on its own shelf — proposing and approving under the policy you set.\n\n```\naw team up\n```\n\nThe blueprint\n\nThe commands above build your team from a first-party blueprint — a versioned set of proven roles you own and evolve on your own shelf.\n\nWhy this exists\n\nA coordinator routes the work, a developer writes the code, a reviewer checks it. Each role needs a clear, stable account of its job.\n\nEvery profile is versioned by digest and every change is signed with your team's [awid](https://awid.ai) identity — so what you start from and evolve is reproducible and trusted.\n\nWhat library gives you\n\nWhat it is\n\nlibrary is built for agents from the ground up: its whole API is part of the aweb protocol, so any agent — or person — can discover and drive it without writing custom code.\n\nCLI-native API\n\nA public manifest maps library's whole API to `aw`\n\ncommands. No integration to write, no SDK to wire up — you just run `aw library`\n\n.\n\nShips agent docs\n\nAn `llms.txt`\n\nand a set of skills ship with library, so any agent that finds it gets readable docs and ready-to-run operations.\n\nVerified by identity\n\nThe manifest is public and pinned by a digest; every call is signed with your team's [awid](https://awid.ai) — auditable and tamper-evident.\n\nIn practice: a person and an agent run the exact same `aw library`\n\ncommands. Because the manifest is machine-readable, an agent discovers and operates library with no custom code.\n\nThe model\n\nThe public catalog holds first-party blueprints — today, aweb.team — that any team can start from. Your shelf is your team's private working set: you adopt profiles onto it, materialize them into runnable agent homes, and improve them under review.\n\nAn agent's job description as a file: mission, instructions, the tools it may use, the actions that need a human's sign-off, and its skills. Versioned by content digest.\n\nThe public, versioned catalog of first-party blueprints — today just `aweb.team`\n\n, a proven set of roles like coordinator, developer, and reviewer any team can start from. Any team can publish into it.\n\nYour team's own copies — started from a blueprint or authored fresh — the working set you edit and own.\n\nCreating or refreshing an agent turns a profile into its runnable home: a composed AGENTS.md, installed skills, and the full profile under `.aw/profile/`\n\n.\n\nAn agent proposes a new version from what it learned; your team reviews and approves, and library mints it — immutably versioned by digest, with the signer recorded.\n\nPull a newer blueprint version's improvements into the parts you have not edited — a per-part merge that never clobbers local work.\n\nFor engineers\n\nThese four properties hold at every version, for every team.\n\n`update-from-source`\n\ntakes upstream blueprint changes only where you haven't edited locally — an existing version is never overwritten.Scopelibrary defines how agents behave — it does not run agents, route messages, or manage compute. v0 has no dashboard and emits no events.\n\nStart from a proven profile, evolve it your way.", "url": "https://wpnews.pro/news/where-teams-of-ai-agents-choose-keep-and-improve-the-profiles-they-run", "canonical_source": "https://library.aweb.ai/", "published_at": "2026-07-10 18:26:13+00:00", "updated_at": "2026-07-10 18:35:10.878092+00:00", "lang": "en", "topics": ["ai-agents", "developer-tools", "ai-infrastructure", "ai-products"], "entities": ["Aweb.ai", "Library", "aw CLI", "tmux", "awid"], "alternates": {"html": "https://wpnews.pro/news/where-teams-of-ai-agents-choose-keep-and-improve-the-profiles-they-run", "markdown": "https://wpnews.pro/news/where-teams-of-ai-agents-choose-keep-and-improve-the-profiles-they-run.md", "text": "https://wpnews.pro/news/where-teams-of-ai-agents-choose-keep-and-improve-the-profiles-they-run.txt", "jsonld": "https://wpnews.pro/news/where-teams-of-ai-agents-choose-keep-and-improve-the-profiles-they-run.jsonld"}}