The open-source workspace for your agents.

Build agents through chat. Improve them with feedback. Share them with your whole team.

Start building Read the docs

Templates

An agent for every job

Start from a working template for engineering, support, sales, knowledge, and operations, then make it yours.

EngineeringCode review agentReviews every pull request and leaves inline comments.

Trigger

On every pull request

Harness

Models

Self-hosted

Skills

review-diffstyle-checkrisk-scan

Tools

GitHubLinear

Use this template

AGENTS.md

# Code review agent

You review pull requests for this repository. Be concise and specific. Never approve your own suggestions.

## When triggered
A pull request is opened or a new commit is pushed to an open PR.

## Steps
1. Read the PR title, description and linked Linear issue
2. Review the full diff, file by file
3. Summarize what changed and why, in 3 sentences max
4. Flag bugs, security issues and perf regressions inline
5. Suggest concrete, minimal fixes with code snippets

## Rules
- Comment only where you would block a merge
- Prefer one high-signal comment over five nitpicks
- Never rewrite whole files, propose diffs
- Style issues: link the guide, do not lecture

## Escalation
Tag the code owner when the diff touches auth, billing or migrations.

How it works

Do the work in chat. Then automate it.

Your agent works with your apps and files. One more message puts it on a schedule.

Work in chatBuild custom agents that share your context and integrations.

2m ago YOU

Check how the new onboarding flow is performing in PostHog and update the onboarding PRD in the workspace with the results.

AGENT▸ Thought for 2s

posthog.QUERY_FUNNEL3 results · 1.2s

workspace.WRITE_FILEonboarding-v2-prd.md · 0.6s

Onboarding completion is 64%, down 4% week over week. Biggest drop-off is step 3, connect data source. I updated onboarding-v2-prd.md with the numbers and a short read.

2m ago · 5.3s · 28.3K tokens · $0.14

Files4

onboarding-v2-prd.mdedited just now

funnel-week-29.csvnew

wiki.md2 days ago

customer-interviews.mdlast week

Ask the agent...

⏎ send · ⇧⏎ newline

AutomateTell it when to run. The same agent keeps doing the job without you.

now YOU

Do this every Monday at 9:00. Flag anything that drops more than 10%.

AGENT

Agent scheduled

On scheduleMondays · 09:00 · flag drops > 10%

Monday 09:00 — ran while you were out

AGENT

workspace.WRITE_FILEonboarding-v2-prd.md · 09:02

FlaggedStep 3 drop-off increased 12% this week. PRD updated. Worth a look before standup.

Mon 09:02 · 41s · 12.1K tokens · $0.06

Files4

onboarding-v2-prd.mdedited Mon 09:02

funnel-week-29.csvnew · Mon 09:01

wiki.md2 days ago

customer-interviews.mdlast week

Ask the agent...

⏎ send · ⇧⏎ newline

Open standards One agent package, any stack Your agent is AGENTS.md, skills, and MCP tools. Swap the harness, model, or runtime without touching a line. When self-hosting locally, you can use your Claude or ChatGPT subscription.

Skills

review-diff classify research-company pull-metrics style-check answer-cited draft-outreach risk-scan build-dashboard

MCPs & tools

Harness

Claude Code pi.dev Codex Soon OpenCode Soon Hermes Soon

Providers

Anthropic OpenAI Ollama xAI OpenRouter Your own gateway

Runtime

Your local machine Agenta Cloud Daytona E2B Soon Vercel Soon

Reliability

Build agents you can trust

Waiting for your approval

Approve & send Edit

Human in the loop Consequential actions wait for you. Approve, edit, or reject before anything ships.

Version every change Prompts, skills, and tools are versioned like code. Roll back to any point, anytime.

lead-research · run #312 6.2s · $0.041 · 5 steps

Trace every run See each step, its cost, and its outcome — and find the exact failure points.

v14 · quality +3

Improve continuously Feedback becomes test cases; every change is evaluated before it ships.

Open source Self-host Agenta on your infrastructure MIT-licensed and yours to run. Keep your agents and data on your own premises — same code as Agenta Cloud, no lock-in.

MIT license Docker Compose Helm charts

Explore on GitHub Self-hosting docs

self-host — with your coding agent

# paste into Claude Code — it walks you through setup and testing $ npx skills add Agenta-AI/agenta-skills > Help me self-host Agenta with its repository. → Agenta running at agenta.acme.com

The open-source workspace for your agents.

Ship agents that actually work