cacardinal/laila
Laila is an operating system for a life run with AI agents. Your world as plain files: domains, memory, hard autonomy lines, background loops. Works on any harness that can read a repo. Private by design; reference implementation on Claude
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 cacardinal/laila/.claude ./rig-laila # 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 cacardinal/laila/.claude/skills/comms .claude/skills/comms $ npx degit cacardinal/laila/.claude/skills/compose .claude/skills/compose $ npx degit cacardinal/laila/.claude/skills/daily-brief .claude/skills/daily-brief $ npx degit cacardinal/laila/.claude/skills/domain-hygiene .claude/skills/domain-hygiene $ npx degit cacardinal/laila/.claude/skills/grill-me .claude/skills/grill-me $ npx degit cacardinal/laila/.claude/skills/idea-deep-dive .claude/skills/idea-deep-dive $ npx degit cacardinal/laila/.claude/skills/laila-headless-conduct .claude/skills/laila-headless-conduct $ npx degit cacardinal/laila/.claude/skills/laila-judgment .claude/skills/laila-judgment $ npx degit cacardinal/laila/.claude/skills/laila-ops-debugging .claude/skills/laila-ops-debugging $ npx degit cacardinal/laila/.claude/skills/level-up .claude/skills/level-up $ npx degit cacardinal/laila/.claude/skills/new-domain .claude/skills/new-domain $ npx degit cacardinal/laila/.claude/skills/os-audit .claude/skills/os-audit $ npx degit cacardinal/laila/.claude/skills/persona .claude/skills/persona $ npx degit cacardinal/laila/.claude/skills/quarterly-refresh .claude/skills/quarterly-refresh $ npx degit cacardinal/laila/.claude/skills/roast .claude/skills/roast $ npx degit cacardinal/laila/.claude/skills/running-brief .claude/skills/running-brief $ npx degit cacardinal/laila/.claude/skills/session-wrap .claude/skills/session-wrap $ npx degit cacardinal/laila/.claude/skills/setup .claude/skills/setup $ npx degit cacardinal/laila/.claude/skills/spec .claude/skills/spec $ npx degit cacardinal/laila/.claude/skills/validate-idea .claude/skills/validate-idea $ npx degit cacardinal/laila/.claude/skills/whats-next .claude/skills/whats-next
$ curl -fsSL --create-dirs -o .claude/agents/comms-triage.md https://raw.githubusercontent.com/cacardinal/laila/main/.claude/agents/comms-triage.md $ curl -fsSL --create-dirs -o .claude/agents/crm-searcher.md https://raw.githubusercontent.com/cacardinal/laila/main/.claude/agents/crm-searcher.md $ curl -fsSL --create-dirs -o .claude/agents/daily-sync-coordinator.md https://raw.githubusercontent.com/cacardinal/laila/main/.claude/agents/daily-sync-coordinator.md $ curl -fsSL --create-dirs -o .claude/agents/decision-logger.md https://raw.githubusercontent.com/cacardinal/laila/main/.claude/agents/decision-logger.md $ curl -fsSL --create-dirs -o .claude/agents/meeting-prep-assembler.md https://raw.githubusercontent.com/cacardinal/laila/main/.claude/agents/meeting-prep-assembler.md $ curl -fsSL --create-dirs -o .claude/agents/research-worker.md https://raw.githubusercontent.com/cacardinal/laila/main/.claude/agents/research-worker.md $ curl -fsSL --create-dirs -o .claude/agents/story-reviewer.md https://raw.githubusercontent.com/cacardinal/laila/main/.claude/agents/story-reviewer.md $ curl -fsSL --create-dirs -o .claude/agents/voice-reviewer.md https://raw.githubusercontent.com/cacardinal/laila/main/.claude/agents/voice-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 *)"
]
}
} Skills (21)
commscomposedaily-briefdomain-hygienegrill-meidea-deep-divelaila-headless-conductlaila-judgmentlaila-ops-debugginglevel-upnew-domainos-auditpersonaquarterly-refreshroastrunning-briefsession-wrapsetupspecvalidate-ideawhats-next
Subagents (8)
| comms-triage model: haiku | Evaluates a communications event (email/text/chat item embedded in the dispatch prompt) against config/autonomy-rules.json and returns a strict JSON tier decision — execute (Tier 1), propose (Tier 3), |
| crm-searcher model: haiku | Local CRM query worker. Use whenever a session needs people, companies, opportunities, tasks, goals, or notes from your CRM (e.g. a self-hosted Twenty instance at localhost:3000) — delegate here so ra |
| daily-sync-coordinator model: sonnet | Cross-domain synthesis for the daily brief. Reads state/strategy.md and every active domain's tracking/status.md, returns max 6 bullets tagged [conflict]/[synergy]/[stale]/[drift] with file citations. |
| decision-logger model: haiku | Extracts decisions from a session summary, meeting debrief, or transcript and appends them to knowledge/decisions/YYYY-MM.md in the established format. Dispatched by the session-wrap workflow and afte |
| meeting-prep-assembler model: sonnet | Meeting prep worker. Use when Alex says "prep me for my call/meeting with X," before any scheduled call with a named contact, or when the daily brief needs prep blurbs for today's meetings. Give it at |
| research-worker model: haiku | Cheap web-research worker for parallel fan-outs. Give it ONE focused question; it runs 2-4 web searches, fetches the best primary sources, and returns a compact cited summary (max 400 words). Dispatch |
| story-reviewer model: sonnet | Project-agnostic QA gate for a completed story pack. The BMAD-style review persona — reads the story file (and, for code-stage, the changed code) and verifies every acceptance criterion is met, the ve |
| voice-reviewer model: sonnet | Reviews any draft written AS Alex (email, text, chat, DM) before it's shown to them. Dispatched by the compose workflow's review gate; also usable standalone ("voice-check this draft"). Give it the dr |
Similar rigs
craigcossairt/trellis
The structure a project grows on: a free starter template for AI-assisted development. One AGENTS.md every tool reads (Claude Code, Cursor, Grok Build, Codex, Gemini CLI, Copilot), TDD + bug-fix methodology, secret guardrails, optional loca
Skill Collector 9.4k tok ·
alirezarezvani/gaios
Open-source AI Operating System (AIOS) blueprint for Claude Code & Codex — turn any agent into your personal second brain + Chief of Staff: skills, self-verifying workflows, a second-brain wiki, and deterministic tools. Fork it for any role
Skill Collector 4.7k tok ·
menkesu/awesome-pm-skills
28 AI-powered PM skills from Lenny's Podcast. Transform Claude Code & Cursor into world-class product partners.
Skill Collector 2.1k tok ·
tammai/ai-sdlc
Claude Code plugin for vibe coding on the AI-native SDLC playbook: intent → design → spec → plan → verify → review → ship. Runs on a Claude subscription, with opinionated stacks (Nuxt/Next + Cloudflare, Go + Postgres, Flutter, Tauri) and de
Orchestrator 3.2k tok ·