Stagent
Free Claude Code plugin that runs long tasks through a state machine you define, so they finish
Editorial take
It is best suited for claude code users who run long, multi-step coding tasks and want enforced review loops and a live view of each run. A free plan is available, with paid plans starting at Free.
What is Stagent?
Stagent is a free Claude Code plugin from Worldstate Labs that runs a task through a state machine you declare in a workflow.json file: stages such as plan, execute, review and QA, the transitions between them, and the artifact each stage must write. A Claude Code stop hook blocks the session from ending until the current stage's report exists with a valid result, and a failed review or QA stage loops back to execution, so the model that wrote the code does not get to declare itself done. In the default cloud mode, a web viewer streams the stage timeline, artifacts and git diff live, sessions can be paused and resumed on another signed-in machine after a git history check, and templates can be published to a hub and run with --flow=cloud://author/name. Local mode keeps everything in the project folder with no network, and /stagent:create turns a plain-English description into a template. Your code runs locally; in cloud mode only stage reports, state and git diffs are synced. (An earlier version of this page described it as a no-code AI agent builder for business process automation, which was incorrect.)
Best for
Claude Code users who run long, multi-step coding tasks and want enforced review loops and a live view of each run
Score breakdown (out of 5)
What you would use it for
- →Running long Claude Code builds that should not stop at almost done
- →Enforcing plan, implement, review and QA loops or strict TDD
- →Watching a Claude Code run from another device
- →Resuming an agent session on a different machine
- →Sharing reusable workflow templates with a team
Pros & Cons
👍 Pros
- ✓Free to use
- ✓Separate review stages stop Claude from marking its own work as done
- ✓Source code stays on your machine; only reports, state and diffs sync
- ✓A session URL lets you or a teammate follow a run from a phone or browser
👎 Cons
- ✗Works only with Claude Code
- ✗Anonymous cloud sessions can be viewed by anyone with the link; sign in for private sessions
- ✗Described as open source, but the GitHub repo has no license file
- ✗The plugin's last release (1.1.2) was in May 2026
- ✗Requires jq, curl and git
Key Features
- ✓ Stages, transitions and required artifacts declared in workflow.json
- ✓ Stop hook that keeps Claude Code running until the stage artifact is valid
- ✓ Review and QA stages that loop back to execution on failure
- ✓ Live browser viewer with stage timeline, artifacts and git diff
- ✓ Pause with /stagent:interrupt and resume on another machine with /stagent:continue
- ✓ Template hub with public, unlisted and private templates
- ✓ /stagent:create builds a template from a plain-English description
- ✓ Local mode with no network
Available on
Integrates with
Stagent Pricing
✅ Stagent has a free plan - no credit card required to start.
Free
- ✓Claude Code plugin, hosted live viewer and template hub
- ✓Requires Claude Code and its own plan or API usage
- ✓Sign in with GitHub, Google or Apple to publish templates and resume sessions
Developer resources
Related Tools
Claude is the best AI assistant for long-document work and research where accuracy matters. Its context window of up to 1M tokens and resistance to sycophancy make it the honest choice when you need a thinking partner rather than an agreeable assistant. Start on the free tier; the Pro plan at $20/month is worth it if you regularly hit the free usage limits, and Max ($100/month) or Ultra ($200/month) suit heavier daily use.
ChatGPT is the most capable all-in-one AI assistant available. Its combination of image generation, voice mode, code execution, web browsing, and a large custom GPT ecosystem makes it the default choice for users who want one tool that handles everything. The tradeoff is that it is rarely the deepest tool on any single task.
Cursor is the most powerful AI code editor available. Its multi-file editing is ahead of plugin-based alternatives, making large-scale refactors practical in ways GitHub Copilot cannot match. At $30/month for Premium (Grok models included, other models billed at API rates) it is a meaningful upgrade for any developer doing more than small isolated changes.
GitHub Copilot is the right starting point for developers adopting AI-assisted coding. It integrates into whatever editor you already use, requires no workflow change, and has a usable free tier. The Pro plan at $10/month is the lowest-cost path to unlimited AI completions, though chat and agent work now draw from a monthly credit allowance; heavy agent users should look at Pro+ ($39) or Max ($100).
This page contains affiliate links. We may earn a commission at no extra cost to you. Learn more.