Aeon
Aeon is an open-source autonomous agent framework that runs self-healing skills on GitHub Actions on a cron schedule.
Aeon earns its keep on recurring chores no one wants to own: PR triage, vuln disclosure to merge, cost and market monitoring. The self-healing loop plus circuit breaker is the part that matters, because a scheduler without failure handling quietly rots. Pass if you want a back-and-forth assistant or sub-minute latency, since cron cadence is not real-time.
Verified 5d ago · liveness 75/100 · cite: rightaichoice.com/tools/aeon
- Developers automating recurring repo work: PR review, triage, changelogs, inbox cleanup
- Security researchers who want vuln-scanner to audit repos, patch and track disclosure to merge
- Crypto teams monitoring tokens, prediction markets, unlocks and on-chain activity on a schedule
- Engineering teams that want a no-ops background agent with self-repair instead of a copilot
- Anyone who wants interactive chat with instant back-and-forth answers
- Non-technical users uncomfortable with GitHub repos, YAML workflows and MCP OAuth setup
- Workflows needing real-time or low-latency responses rather than scheduled runs
We scan live Reddit threads, YouTube comments, X posts, G2 reviews and other communities — and hand you an honest verdict in under a minute.
- Honest verdict, not marketing
- Real pros & cons from real users
- Attributed quotes with receipts
3 free scans · no card needed
Skip Aeon if your work needs interactive back-and-forth or real-time responses rather than scheduled GitHub Actions runs, or if you don't want to own a forked repo, YAML workflows and MCP OAuth credentials yourself.
Private instances run on your own GitHub Actions minutes — 2,000 a month free on a free account, then billed by GitHub, and a 5-minute scheduler tick plus skill runs can eat that faster than you expect
Aeon is an MIT-licensed open-source framework — the framework itself costs nothing, so your real spend is GitHub Actions minutes plus the model and MCP calls each run makes. That puts the floor well below per-seat agent subscriptions for solo developers and small teams whose work already lives in GitHub. It rises above those subscriptions if you enable many frequent skills on a private instance, because every run spends tokens you pay for directly.
In short
Aeon — Aeon is an open-source autonomous agent framework that runs self-healing skills on GitHub Actions on a cron schedule. Best for Developers automating recurring repo work: PR review, triage, changelogs, inbox cleanup, Security researchers who want vuln-scanner to audit repos, patch and track disclosure to merge, Crypto teams monitoring tokens, prediction markets, unlocks and on-chain activity on a schedule. Free to use.
What's new in Aeon
Checked 4 days agoAcross the latest 10 updates: 4 feature updates, 1 launch, 2 changelog entries and 3 community discussions.
compute-resell: free compute sold at the clearing price
Aeon's compute-resell skill sells idle provider compute for USDC, priced against sellers that can actually serve buyers.
Hosted MCP server becomes default integration path; Haiku 5.5 tier
Hosted MCP server at aeon.fun/connect/mcp is now the documented default; Claude Haiku 5.5 replaces Haiku 4.5 in the haiku tier.
Anatomy skill: interactive isometric figures; catalog hits 86 skills
New anatomy dev skill turns a one-line brief into an interactive isometric SVG figure. Catalog now 86 skills.
Ten Skills That Keep an Unsupervised AI Agent Honest
Ten Aeon skills mapped around one pattern: each refuses to lose a receipt, a verdict, or a human gate. Skill files free to take.
Higgsfield skill: three modes, 100+ models, two outputs
Aeon's higgsfield skill exposes three modes over 100+ models, with two outputs per run.
Arc Studio skill: one dev session split across cron runs
The arc-studio skill drives Circle Arc Studio headless: start an Arc testnet turn, poll it later, get pinged if needed.
Skill Packs, Cult OS, and $AEON
Skills move between aeon forks as free, scanned community packs. Work is hired per call over x402 on Cult OS.
Aeon Connect: start your agent from the browser
Aeon Connect is live at aeon.fun/connect. Sign in with GitHub, connect a model, pick skills. The agent still lives in your repo.
aeon vs OpenAI Dots: open-source agent or $500-a-month digital employee?
Comparisons between Aeon's repo-based always-on agent and OpenAI Dots' cloud-hosted always-on agents.
Aeon vs Instinct: the agent that reads your repo vs the one that reads your screen
Side-by-side with Instinct on access, credentials, training and cost. Aeon sees only what you hand it a secret for.
What people actually say about Aeon — is it worth it?
We scanned public community sources for Aeon on Sep 24, 2026 and could not establish that the discussion we found is about this tool rather than something else sharing its name. Only 3 of the posts we fetched could be positively tied to Aeon. Rather than publish a sentiment score built on the wrong subject, we publish nothing here and re-run the scan.
Viability Score
How well maintained and how widely used is Aeon? Built from what the vendor actually publishes (docs, changelog, tutorials, integrations, pricing), whether the site is live, and how much real users discuss it. How we calculate this
Last calculated: October 2026
How we score →Key Features
- Configure-once autonomous agents that run on GitHub Actions on a cron schedule
- 86 built-in skills across Dev, Research, Crypto, Security, Social and Productivity
- Self-healing loop where skill-repair diagnoses and patches degraded skills
- Circuit breaker parks repeatedly failing skills to stop wasting Actions minutes
- Cron scheduler ticks every 5 minutes and dispatches due skills from aeon.yml
- Fleet control spawns instances, forks fleets and tracks divergence via heartbeat
- Reactive Telegram, Discord and Slack bots with inline buttons and slash commands
- Agent wallet with x402, USDC and Circle for autonomous API payments
- One-click MCP OAuth for Base, Robinhood and Executor servers
- Vuln Scanner audits trending repos, lands fixes and runs disclosure to merge
- Deploy Uni Hook generates, simulates, audits and deploys a Uniswap v4 hook plus test pool
- Self-improvement: the agent authors new skills as Markdown and Create Skill ships them
- Sigstore attestation signs every run for verifiable provenance
- Memory persists context across runs via cron-state.json, MEMORY.md and CLAUDE.md
- One-command setup (./aeon init) with a credential manifest and 266-check drift test
About Aeon
Aeon takes a different angle on AI agents: instead of a chat window you babysit, it is a configure-once framework that lives inside your own GitHub repo. You fork or template it, write direction into SOUL.md and STRATEGY.md, enable the skills you want in aeon.yml, and GitHub Actions dispatches them on a cron tick every five minutes. No server, no Dockerfile, no daemon to keep alive. The catalog now runs 86 skills across Dev, Research, Crypto, Security, Social and Productivity. That spans PR Review and PR Triage, Vuln Scanner and Vuln Tracker, Changelog, Seo Audit, Spend Watch, GitHub Monitor, Inbox Triage, Token Movers, Unlock Monitor, Monitor Polymarket, Plus Fleet Control, Spawn Instance, Skill Repair and Create Skill. Runs are scored, degraded skills get diagnosed and patched by skill-repair, and a circuit breaker parks anything that keeps failing so it stops burning Actions minutes. Setup is now a single command. Recent releases bring ./aeon init with a credential manifest held honest by a 266-check drift test, one Connect modal in the dashboard, plain-language failed-run reasons, and a phone layout. Model pickers default to current releases rather than old pins. October 2026 shipped a security, sandbox, scheduler, state and runtime audit, moved every harness CLI to its latest release with CI smoke tests, and now signs runs with Sigstore attestation. It suits engineers automating unattended recurring work, security researchers running end-to-end vulnerability disclosure, and crypto teams tracking tokens, prediction markets and unlocks on a schedule. You own the repo, keys and logs, and you carry the model and Actions-minute costs. Alternatives keep you in the driver's seat interactively; Aeon is built to run when you are not looking.
Behind the Verdict
We'd reach for Aeon when the work is repetitive, scheduled and verifiable. A nightly Seo Audit, a weekly Changelog, an Inbox Triage that surfaces aging vuln PRs before they rot, a Spend Watch that reads cloud costs across Neon, Vercel, Railway and GitHub. These are chores that a copilot makes you do manually every time. The self-healing design is what separates it from a cron job you wrote yourself. Runs get scored, skill-repair patches degraded skills, and the circuit breaker stops repeat failures. If you have ever maintained a scheduled script that silently broke in March and nobody noticed until June, that distinction is the whole product. The October 2026 cycle is worth noting if you evaluated earlier: a security, sandbox, scheduler, state and runtime audit landed, harness CLIs moved to current releases with CI smoke tests, the community pack registry was trimmed to eight live packs, and the install-from-atrium path was removed. Model pickers now default to current releases rather than whatever was pinned at fork time, which matters because a stale default is a silent quality drop. Where it bites: this is not an interactive tool. There is no chatting your way out of a bad run, and responses arrive on cron cadence, not in seconds. Setup assumes comfort with GitHub repos, YAML and MCP OAuth, even with ./aeon init smoothing the first pass. You also inherit the maintenance of a fork, and you pay the model plus Actions-minute bill yourself. Compared with Claude Code or Codex harnesses, Aeon trades steering for autonomy. Those keep a human in the loop and are better for exploratory, judgement-heavy work. Aeon is better when the task is known, repeatable, and should just happen. The other comparison the project invites is OpenAI Dots, a $500-a-month hosted digital
Researching Aeon? Get your full AI stack in 60 seconds.
Free, no signup — tell us your goal and get tools matched to your budget & existing stack.
Real-world workflow fit
Concrete scenarios for the personas Aeon actually fits — and what changes day-one when you adopt it.
You run ./aeon init from the terminal, connect a Claude subscription token, enable PR Review and PR Triage in aeon.yml, and let the 5-minute scheduler tick dispatch them. Each run is scored 1–5 by Claude Haiku, and skill-repair patches a skill that starts failing.
Outcome: External PRs get a first-touch verdict and label without you opening the repo, and a skill that breaks gets diagnosed instead of silently going dark.
You enable Vuln Scanner and Vuln Tracker on a public instance so Actions minutes are free, and let the disclosure pipeline open PRs on repos it finds issues in. Sigstore attestation signs each run so provenance is checkable.
Outcome: Vulnerabilities get audited and fixes land as PRs, with a lifecycle poll tracking each one from disclosure to merge.
You wire Base MCP and Robinhood MCP via one-click OAuth, enable Token Movers, Unlock Monitor and Monitor Polymarket, and point the agent wallet at x402 for any paid data calls it needs to make autonomously.
Outcome: Token movers, unlock supply pressure and prediction-market price moves arrive on a schedule, and the agent pays for its own API calls rather than stalling for approval.
Use Cases
- Automate daily PR review, first-touch triage, and changelog generation on your GitHub repos
- Scan trending or open-source agent repos for vulnerabilities and carry disclosure to merge
- Monitor DeFi markets and get scheduled summaries of token movers, unlocks and sentiment
- Generate daily research digests from papers, Hacker News and Reddit
- Run a weekly on-page and technical SEO audit of every page on a site
- Track cloud spend across Neon, Vercel, Railway and GitHub with Spend Watch
- Monitor Polymarket and Kalshi prediction markets for 24h price moves
- Triage GitHub notification inboxes and surface aging vuln PRs daily
Models Under the Hood
as of 2026-10-17
Limitations
- Aeon is an MIT-licensed, open-source framework that runs unattended on GitHub Actions on a cron schedule rather than as an interactive real-time tool.
- You own the repo, keys and memory, and you only pay for the model and API keys you connect; public repositories get unlimited free Actions minutes, though memory and run logs in a public repo are public.
- Headless runs are bounded by a 50-minute job cap, and each run is scored 1-5 by Claude Haiku, with repeatedly failing skills diagnosed and patched autonomously.
as of 2026-10-04
Verification history
We have re-verified Aeon 10 times since . Each pass re-reads the vendor's own pages and re-checks every listed field against that evidence; passes where nothing had changed are marked as such.
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
- — re-verified summary, description, our verdict, our analysis, pricing model, pricing tiers, features, integrations, who it suits, who should skip it
Showing the 6 most recent of 10 verification passes.
Free to cite with attribution — this page re-verifies continuously.
12-month cost
Project the real annual outlay, including the implied monthly cost when only an annual tier is published.
Vendor list price only. Add-on usage, seat overages, and contract minimums are surfaced under Hidden costs & gotchas.
Plans compared
For each published Aeon tier: who it actually fits, and what it adds vs. the previous tier. Cross-reference the cost calculator above for projected annual outlay.
Free
$0/mo
Ideal for
Developers, security researchers or crypto operators who want an always-on agent inside a repo they already own
What this tier adds
Starting tier: MIT-licensed framework at $0/mo, with public repos getting unlimited free GitHub Actions minutes and private ones using your own minutes
Where the pricing makes sense
The company stage and team size where Aeon's pricing actually pencils out — and where peers do it cheaper.
Aeon is an MIT-licensed open-source framework — the framework itself costs nothing, so your real spend is GitHub Actions minutes plus the model and MCP calls each run makes. That puts the floor well below per-seat agent subscriptions for solo developers and small teams whose work already lives in GitHub. It rises above those subscriptions if you enable many frequent skills on a private instance, because every run spends tokens you pay for directly.
Setup time & first value
How long it actually takes to get something useful out of Aeon — broken out by persona, not the marketing-page minute.
Browser path (Aeon Connect): about five minutes — sign in with GitHub, name the repo, install the GitHub App on that repo, paste a model credential, pick skills. Terminal path: git clone then ./aeon init, which creates your repo from the template, pins gh, turns on Actions, connects a model and links Telegram; needs Node.js 20+ and gh auth login. Coding-agent path: paste one line into Claude
Switching to or from Aeon
How to bring data in from common predecessors and how to get it back out — written for the switcher, not the buyer.
- →From cron-plus-scripts: move each script behind a skill in aeon.yml and let the 5-minute scheduler tick dispatch it, gaining scoring, skill-repair and a circuit breaker
- →From a hand-run PR review habit: enable PR Review and PR Triage and let runs execute on their schedule instead of on demand
- →From Claude Code or another harness: keep the harness and point the agent at your repo — Aeon works with Claude Code, Codex, Cursor, Hermes, OpenClaw and GLM
- →From a self-hosted agent daemon: delete the server and Dockerfile and run the same work on GitHub Actions from a repo you own
- ↗To plain cron plus scripts: extract each skill's logic from the repo and run it on your own scheduler, losing run scoring, skill-repair and the circuit breaker
- ↗To an interactive copilot: move recurring work back to on-demand sessions in Claude Code or Codex, accepting that nothing runs while you are away
- ↗To a hosted agent product: hand over the runtime and credentials in exchange for vendor support, giving up the fork you own
Integrations
Resources & Guides
Tutorials & Learning
YouTube returned 6 videos for “Aeon”, and we withheld 6: 6 could not be judged, because “Aeon” is a single word that other videos use for other things. We are showing none, because we could not prove any of them are about Aeon.
Official links
Tools that pair well with Aeon
Common stack mates teams adopt alongside Aeon, with the specific reason each pairing earns its keep.
AgentScope
AgentScope is an open-source Python framework from Alibaba for building distributed multi-agent AI systems.
Hermes Desktop
Hermes Desktop is a free, open-source desktop GUI for the Hermes Agent — a self-hosted AI agent with a built-in learning loop that turns
Antigravity (Google)
Google Antigravity runs parallel local coding agents across real codebases from $0/month.
Featured Head-to-Head Comparisons
Aeon vs Spider Cloud
If you need a set-and-forget agent to handle recurring dev, security, or crypto tasks, choose Aeon (free and open-source). If you need fast, reliable web data for AI agents or RAG, choose Spider Cloud (pay-as-you-go, Rust-powered). They solve different problems—Aeon for long-running background automation, Spider Cloud for on-demand crawling.
Aeon vs Temporal Ai
Temporal is the powerhouse for teams needing reliable, durable execution of complex workflows with human-in-the-loop and full visibility. Aeon is for hackers who want a zero-maintenance background agent that runs forever on GitHub Actions. Choose Temporal if you orchestrate mission-critical AI agents; choose Aeon if you want a set-and-forget bot for recurring tasks.
Aeon vs Presto Voice
If you operate a QSR drive-thru and want proven revenue lift through voice AI, Presto Voice is the clear choice—but it requires budget and vendor lock-in. If you need a free, developer-run agent to automate dev tasks, security scans, or crypto monitoring, Aeon delivers unmatched autonomy with zero infrastructure cost. The two tools serve completely different verticals; your decision hinges on whether you are a restaurant operator or a technical automation builder.
Alternatives to Aeon
View allAgentScope
AgentScope is an open-source Python framework from Alibaba for building distributed multi-agent AI systems.
Hermes Desktop
Hermes Desktop is a free, open-source desktop GUI for the Hermes Agent — a self-hosted AI agent with a built-in learning loop that turns
Antigravity (Google)
Google Antigravity runs parallel local coding agents across real codebases from $0/month.
Frequently Asked Questions
Used Aeon? Help shape our editorial sentiment research.