assay CLI
CLI checks a web page in Chromium: finds every control, uses it and reports what broke. No tests to write and no LLM.
- 110 ★ on the repository
- Last commit
- MIT
Extended access
Why this level
- Launches Chromium and clicks every control on a page; on pages with side effects this can change data.
Install
- Works in
In your terminal, with SkillFoxx CLI
npx skillfoxx add cli/assayDetects the agents on your machine, shows the access level and pins the version.
Other ways to install
Access and checks
How we reviewExtended accessThe tool runs code or makes network calls and only reads external data.
- Launches Chromium and clicks every control on a page; on pages with side effects this can change data.
This is third-party code. Review the repository files before installing.
What it does
The `assay ./my-app` command opens the page in Chromium, finds every control on it, tries them all and reports what went wrong. Checks are built from the controls it finds, so there are no tests to write. No LLM or API key is involved, so a run is free and gives the same result every time. A skill for 15 agent harnesses ships with the CLI: it tells the agent to check each page it writes and end its reply with a result line. The plugin adds a hook that runs the check after every turn that changed a page. Repeated findings are grouped, with a reference such as `same finding as C003` instead of repeated text.
Who it is for. Developers who have an agent build pages and want an automatic check that buttons and fields actually work.
Good fit when
- An agent built a static page and you need to confirm dragging and buttons still work.
- A team wants a hook to check changes after each turn in Claude Code or Codex.
- A script or agent needs a short one-line result of a run.
- A check report needs screenshots in an HTML file.
Not a fit when
- You need scenarios with custom business logic and data checks; assay writes no tests.
- The page cannot be opened in Chromium, for example behind a closed login.
- You need a code review rather than interface behavior.
Example request
Check the page in ./todo with assay and tell me which checks failed.Limitations
Python 3.10 or newer is required, and Chromium is downloaded on the first run. The tool checks a page from a local folder, and the agent does not change code based on results unless you ask. The hook is not available everywhere: it exists in Claude Code, Codex, Grok Build CLI, Devin CLI, Factory Droid, GitHub Copilot CLI and DeepSeek Harness, while other harnesses get only the skill. The exit code is non-zero if any check fails, which matters in CI. Checks are generated automatically, so they do not cover deep application logic.
How to disable. Remove the CLI with `pip uninstall assay-ui` and disable the plugin or skill in your agent harness.
FAQ
How do I install assay?
Run `pip install assay-ui`, which needs Python 3.10 or newer. `uv tool install assay-ui` and `pipx install assay-ui` also work.
Does it need an API key or a model?
No. Checks run without an LLM: no keys, no cost, and the result repeats from run to run.
How do I connect assay to Claude Code?
Install the CLI, then run `/plugin marketplace add awss1i/assay` and `/plugin install assay@assay`. The plugin adds the skill and a post-turn check hook.
How do I get a report with screenshots?
Run `assay ./my-app --report out.html` to write an HTML report with screenshots alongside console output. The `--one-line` flag prints a short summary for scripts.
Which harnesses does the skill support?
The skill supports 15 harnesses, including Claude Code, Codex App and CLI, Cursor, Gemini CLI, opencode, GitHub Copilot CLI, Hermes Agent and Kimi Code.
Discussion
No questions or reviews yet.
Related
CLI toolsA skill that tests your agent's code in a real browser: reads your changes, builds a test plan and runs it through Playwright with your own cookies
Microsoft’s CLI controls real terminal sessions, checks visible text and records test runs for investigating failures.
kogiQA MCP gives an agent a browser that finds page elements without selectors, for debugging web apps and writing e2e tests
A CLI for testing and red teaming LLM applications: compares models, runs automated CI checks and finds vulnerabilities
Microsoft's official CLI runs graded trials for skills and agents, checking the evidence behind a result, not just pass or fail.
One binary for agent browser automation: a CLI with browser and check skills, an MCP server and clients for JS, Python and Java
README in short
The README covers installation through pip, uv or pipx, running the `assay` command, report flags and connecting the skill or plugin through the harnesses page.