CONFIGURE ONCE FORGET FOREVER

eaon - the most autonomous agent framework. Give it a direction and it leverages81 skills to get the work done. No approval loops, no babysitting. Open source, MIT licensed.

The most autonomous agent framework

No approval loops, no babysitting.

RUN NOW733 SEE SKILLS

read aeon.fun/skills/aeon.md ...

Works with

Claude CodeCodexHermesOpenClaw& more

Used by engineers at

• Harvard](https://www.harvard.edu/)
• Deloitte](https://www.deloitte.com/)
• Accenture](https://www.accenture.com/)
• GitHub](https://github.com/)
• Microsoft](https://www.microsoft.com/)
• Google](https://www.google.com/)
• Nvidia](https://www.nvidia.com/)
• SpaceX](https://www.spacex.com/)
• Anthropic](https://www.anthropic.com/)
• OpenAI](https://openai.com/)

01234567890123456789012345678900123456789012345678901234567890SKILLSZERO BABYSITTING

• RESEARCH
• MONITOR
• REVIEW
• DEPLOY
• BRIEF
• SHIP
• SELF-HEAL

Core1 / 81

• Aeon Update Pull framework updates from the upstream Aeon repo into this instance…
• Auto Merge Automatically merge open PRs that have passing CI, no blocking…
• Auto Workflow Two-mode aeon.yml workflow builder - analyze inspects URLs and emits…
• Fleet Control Operate managed Aeon instances from memory/instances.json -…
• Fork Fleet Fork divergence monitor - tracks where the fleet's active forks…
• Heartbeat Ambient fleet-health check that surfaces anything worth attention…
• Memory Flush Promote important recent log entries into MEMORY.md and prune stale…
• Narrative Convergence Cross-skill signal detector - finds entities or themes surfaced…
• Shiplog Recap of everything shipped since the last run - cross-repo PRs,…
• Soul Builder Build a SOUL from an X handle - read a wide sample of a public X…
• Spawn Instance Clone this Aeon agent into a new GitHub repo - fork, configure…

Most autonomous

Most agent tools are interactive - useful for hands-on work, less so for the long tail of recurring tasks: priority briefs, market monitoring, PR reviews, research digests, security scans. aeon owns that tail and runs it unattended.

Self-healing

Every skill output is scored 1–5 by Haiku after each run. Flags like api_error,stale_data, and rate_limited live in a rolling 30-run history.heartbeat audits the fleet daily; if a skill fails 3× in a row,skill-repair diagnoses and patches it.

Zero infra

A Next.js dashboard plus a set of GitHub Actions workflows. Fork the repo, run./aeon, authenticate, pick skills, push. Actions handles scheduling, execution, and notifications via Telegram, Discord, Slack, or Email. Public repo = unlimited free minutes.

STRATEGY.md

• North-star metric

Weekly active teams - not signups.

• Priorities

1. Reliability over features
2. Cut onboarding friction
3. Earn trust in public

• Audience

Technical founders, short on time.

• Never

No growth hacks. No unverified claims.

1. 01 · Aligns Every skill reads your voice and your intent, and lets them break ties - what to prioritise, what to skip, when to bother you.

2. 02 · Builds Chasing a metric it cannot measure yet? aeon writes the missing skill and asks you for the API keys it needs to pull the data.

3. 03 · Self-evolves Run after run it reprioritises the fleet and tunes its own skills - compounding toward your number with no new instruction.