awebai/aweb
Communication for AI agents: stable identity, durable mail and chat, and wake-up events across sessions, runtimes, machines, and organizations. MIT, self-hostable.
ARCHETYPE
Orchestrator
A bench of specialised subagents. The main agent mostly delegates.
Copy this rig
# review before running: this installs third-party code
$ npx degit awebai/aweb/.claude ./rig-aweb # inspect, then merge into .claude/ MCP servers are added to Claude Code at local scope; env vars are shown as YOUR_… placeholders — we never store values. Files are fetched with degit into a separate folder so you can review before merging.
$ npx degit awebai/aweb/.claude/skills/code-review-discipline .claude/skills/code-review-discipline $ npx degit awebai/aweb/.claude/skills/cross-repo-change .claude/skills/cross-repo-change $ npx degit awebai/aweb/.claude/skills/deploy-awid-site .claude/skills/deploy-awid-site $ npx degit awebai/aweb/.claude/skills/production-awid-db .claude/skills/production-awid-db $ npx degit awebai/aweb/.claude/skills/release .claude/skills/release $ npx degit awebai/aweb/channel/skills/configure .claude/skills/configure $ npx degit awebai/aweb/naapp/folio/.agents/skills/retire-instance .claude/skills/retire-instance $ npx degit awebai/aweb/naapp/folio/.agents/skills/self-maintenance .claude/skills/self-maintenance $ npx degit awebai/aweb/naapp/folio/.agents/skills/spawn-instance .claude/skills/spawn-instance $ npx degit awebai/aweb/naapp/folio/skills/agent-first-app .claude/skills/agent-first-app $ npx degit awebai/aweb/naapp/folio/skills/byot-e2e-validation .claude/skills/byot-e2e-validation $ npx degit awebai/aweb/naapp/folio/skills/create-from-template .claude/skills/create-from-template $ npx degit awebai/aweb/naapp/folio/skills/present-to-human .claude/skills/present-to-human $ npx degit awebai/aweb/naapp/folio/skills/set-theme .claude/skills/set-theme $ npx degit awebai/aweb/naapp/folio/skills/team-cert-verification .claude/skills/team-cert-verification $ npx degit awebai/aweb/oats/.agents/capabilities/owned/aweb-tasks/skills/aweb-coordination .claude/skills/aweb-coordination $ npx degit awebai/aweb/packages/codex-plugin/skills/aweb-bootstrap .claude/skills/aweb-bootstrap $ npx degit awebai/aweb/packages/codex-plugin/skills/aweb-identity .claude/skills/aweb-identity $ npx degit awebai/aweb/packages/codex-plugin/skills/aweb-messaging .claude/skills/aweb-messaging $ npx degit awebai/aweb/packages/codex-plugin/skills/aweb-team-membership .claude/skills/aweb-team-membership $ npx degit awebai/aweb/test-<redacted>-home-created/developer/.aw/profile/skills/implement .claude/skills/implement $ npx degit awebai/aweb/test-<redacted>-home-legacy/coordinator/skills/coordinate .claude/skills/coordinate $ npx degit awebai/aweb/test-<redacted>-home-legacy/reviewer/skills/review .claude/skills/review
$ curl -fsSL --create-dirs -o .claude/agents/2026-07-02-agent-runtime-launch-written-once.md https://raw.githubusercontent.<redacted>-07-02-agent-runtime-launch-written-once.md $ curl -fsSL --create-dirs -o .claude/agents/2026-07-03-agent-home-work-isolation.md https://raw.githubusercontent.<redacted>-07-03-agent-home-work-isolation.md $ curl -fsSL --create-dirs -o .claude/agents/handoff-2026-07-26.md https://raw.githubusercontent.<redacted>-2026-07-26.md $ curl -fsSL --create-dirs -o .claude/agents/handoff-2026-07-27.md https://raw.githubusercontent.<redacted>-2026-07-27.md $ curl -fsSL --create-dirs -o .claude/agents/MEMORY.md https://raw.githubusercontent.<redacted>.md $ curl -fsSL --create-dirs -o .claude/agents/app-emit-path-binding-canonical.md https://raw.githubusercontent.<redacted>-emit-path-binding-canonical.md $ curl -fsSL --create-dirs -o .claude/agents/aweb-deploy-topology.md https://raw.githubusercontent.<redacted>-deploy-topology.md $ curl -fsSL --create-dirs -o .claude/agents/awid-cert-controller-only-revoke.md https://raw.githubusercontent.<redacted>-cert-controller-only-revoke.md $ curl -fsSL --create-dirs -o .claude/agents/correctness-over-momentum.md https://raw.githubusercontent.<redacted>-over-momentum.md $ curl -fsSL --create-dirs -o .claude/agents/dont-assert-business-facts-messy-isnt-unneeded.md https://raw.githubusercontent.<redacted>-assert-business-facts-messy-isnt-unneeded.md $ curl -fsSL --create-dirs -o .claude/agents/dont-reopen-converged-decisions.md https://raw.githubusercontent.<redacted>-reopen-converged-decisions.md $ curl -fsSL --create-dirs -o .claude/agents/lead-drive-to-completion.md https://raw.githubusercontent.<redacted>-drive-to-completion.md $ curl -fsSL --create-dirs -o .claude/agents/regression-attribution-reproduce-read-actual-error.md https://raw.githubusercontent.<redacted>-attribution-reproduce-read-actual-error.md $ curl -fsSL --create-dirs -o .claude/agents/review-completeness-every-site-every-dimension.md https://raw.githubusercontent.<redacted>-completeness-every-site-every-dimension.md $ curl -fsSL --create-dirs -o .claude/agents/run-aw-only-from-instance-home.md https://raw.githubusercontent.<redacted>-aw-only-from-instance-home.md $ curl -fsSL --create-dirs -o .claude/agents/security-review-detector-narrowing-check-wrapped-forms.md https://raw.githubusercontent.<redacted>-review-detector-narrowing-check-wrapped-forms.md $ curl -fsSL --create-dirs -o .claude/agents/verification-method-from-aajc.md https://raw.githubusercontent.<redacted>-method-from-aajc.md $ curl -fsSL --create-dirs -o .claude/agents/log.md https://raw.githubusercontent.<redacted>.md $ curl -fsSL --create-dirs -o .claude/agents/operating-continuity.md https://raw.githubusercontent.<redacted>-continuity.md $ curl -fsSL --create-dirs -o .claude/agents/instructions.md https://raw.githubusercontent.com/awebai/aweb/main/agents/instructions.md $ curl -fsSL --create-dirs -o .claude/agents/coordinator.md https://raw.githubusercontent.com/awebai/aweb/main/agents/roles/coordinator.md $ curl -fsSL --create-dirs -o .claude/agents/developer.md https://raw.githubusercontent.com/awebai/aweb/main/agents/roles/developer.md $ curl -fsSL --create-dirs -o .claude/agents/reviewer.md https://raw.githubusercontent.com/awebai/aweb/main/agents/roles/reviewer.md
This rig commits no guardrails. Here is the community baseline instead — the deny/ask rules most often found across all 7,204 rigs:
{
"permissions": {
"deny": [
"Read(./.env)",
"Read(**/.env)",
"Read(~/.ssh/**)",
"Bash(rm -rf *)",
"Read(**/*.pem)",
"Bash(rm -rf /)",
"Bash(git push --force:*)",
"Bash(sudo *)",
"Read(.env)",
"Bash(rm -rf /*)",
"Read(./.env.*)",
"Read(~/.aws/**)",
"Bash(git push --force*)",
"Bash(rm -rf:*)",
"Read(**/*.key)",
"Read(**/.env.*)",
"Bash(sudo:*)",
"Bash(git reset --hard*)",
"Bash(git reset --hard:*)",
"Read(.env.*)"
],
"ask": [
"Bash(git push:*)",
"Bash(git push *)",
"Bash(git commit:*)",
"Bash(rm *)",
"Bash(rm:*)",
"Bash(git rebase *)",
"Bash(wget *)",
"Bash(npm publish:*)",
"Bash(git commit *)",
"Bash(gh pr merge *)"
]
}
} MCP servers (1)
| server | source | est. tokens |
|---|---|---|
| aweb-channel | local / custom | 2.5k |
Skills (23)
code-review-disciplinecross-repo-changedeploy-awid-siteproduction-awid-dbreleaseaw:configureretire-instanceself-maintenancespawn-instanceagent-first-appbyot-e2e-validationcreate-from-templatepresent-to-humanset-themeteam-cert-verificationaweb-coordinationaweb-bootstrapaweb-identityaweb-messagingaweb-team-membershipimplementcoordinatereview
Subagents (23)
| 2026-07-02-agent-runtime-launch-written-once | — |
| 2026-07-03-agent-home-work-isolation | — |
| handoff-2026-07-26 | — |
| handoff-2026-07-27 | — |
| MEMORY | — |
| app-emit-path-binding-canonical | App-emit signed-path binding is mount-dependent and CANONICAL-LOCKED — hosted (app.aweb.ai, ac mounts aweb under /api) = /api/v1/events/app; the frozen conformance vector's /v1/events/app is the STAND |
| aweb-deploy-topology | app.aweb.ai IS ac's deployment; the aweb server is bundled into ac's release image FROM SOURCE, not a standalone service. aweb changes (e.g. m3.x) ship via an ac v* release — there is no separate aweb |
| awid-cert-controller-only-revoke | awid team-certificate model is deliberately controller-only for add AND revoke. The issuer_did_key column + revoke-broadening (revoke-by-issuer-if-member OR holder) was investigated end-to-end and SCR |
| correctness-over-momentum | Optimize for getting it RIGHT, not for activity/momentum. Don't manufacture busywork to avoid idle; deliberate depth on the hard design beats keeping everyone busy. "Drive to completion" means drive t |
| dont-assert-business-facts-messy-isnt-unneeded | Don't assert product/business/roadmap facts (pricing, are-we-charging, what's-needed-soon, which features matter) as known — verify or defer to Juan. And "messy/unproven/sprawl" != "unneeded": clean+c |
| dont-reopen-converged-decisions | Once the team has converged AND committed on a functionally-fine choice, freeze it — don't re-open to optimize a marginal detail. A coordinator's flip-flop costs more thrash (reversals, blocked agents |
| lead-drive-to-completion | As coordinator, take charge and drive work to done; decide what is yours; do not defer coordination to the human or narrate option-menus. |
| regression-attribution-reproduce-read-actual-error | When attributing a regression, REPRODUCE it and read the LITERAL failing error before committing a fix. A bisect proves which DELTA broke it, not which COMMIT — suspect-by-name attribution is a guess. |
| review-completeness-every-site-every-dimension | When gating a fix, verify the fix's OWN stated purpose holds at EVERY affected site, and enumerate ALL completeness dimensions before ACK — don't anchor on the one dimension that's top of mind. I miss |
| run-aw-only-from-instance-home | Run `aw` ONLY from this instance's home dir; never batch git (repo root) + aw in one shell — the repo root is a different identity (grace), so aw there sends as the wrong agent and "agent not found" l |
| security-review-detector-narrowing-check-wrapped-forms | When security-reviewing a change that narrows a broad substring scanner into structured regex detectors, enumerate the SERIALIZATION-WRAPPED forms (JSON/YAML quoted keys, quoted headers), not just bar |
| verification-method-from-aajc | — |
| log | — |
| operating-continuity | The OATS home changes execution layout without replacing the team's current policy or historical record. |
| instructions | — |
| coordinator | — |
| developer | — |
| reviewer | — |
Similar rigs
yanis-vroland/fieldops
Plateforme fictive de maintenance industrielle avec IA intégrée : API NestJS, app Flutter, RAG, MCP et agents. Projet portfolio de développement agentique.
Fort Knox 2.0k tok ·
pruthvidarji1993/ai-automation-jira
AI-assisted dev workflow: Jira ticket in, reviewed PR out. Claude Code plugin + GitHub Copilot prompts with human approval gates (plan, diff).
Pragmatist 2.0k tok ·
yoshpy-dev/ralph
Claude Code + Codex harness engineering — scaffold, upgrade, and run opinionated agent harnesses with deterministic hooks, evidence-backed review pipelines, and an autonomous multi-seat org runtime.
YOLO Cowboy 4.7k tok ·
anujg21/claude-engineering-os
A repository template that gives Claude Code a working engineering method: phased workflow, path-scoped standards, independent review agents, and hooks that block dangerous commands.
Skill Collector 3.1k tok ·