$ loading_
Run policy-gated, local-first Chromium automation with traceable snapshots and telemetry.
Copy the install command and let the AI configure it · recommended for beginners
No copy-paste install info for "SnapRef" yet — see the docs or source repo.
Use SnapRef to start an isolated Chromium session, run a script to open the login page, enter username and password, click the login button, capture a snapshot at each step, and then return execution logs with telemetry receipts.
Return step-by-step snapshots, success status for each action, and a complete telemetry receipt so you can verify the login flow execution.
Open the target page with SnapRef, save a snapshot and a revision-bound ref. Then in a new session, reproduce the same page state using that ref and compare the two snapshots.
Provide a reusable revision-bound reference, comparison results between both snapshots, and whether the page state differs.
In policy-gated mode, use typed actuation to enter a search query step by step and submit it. Return a snapshot and telemetry receipt for every action.
Return per-action snapshots and telemetry receipts to support auditing of automated behavior under policy control.
After frontend changes, developers can replay key interaction paths in an isolated local Chromium session and verify behavior stability through snapshots and revision references.
Operations teams can run scripted browser checks under policy constraints and use telemetry receipts to record outcomes for each run.
SnapRef is an MCP tool that provides policy-gated, local-first browser automation, supporting isolated Chromium sessions, snapshots, revision-bound refs, typed actuation, script execution, and telemetry receipts.
The provided material does not include installation steps. Please check the source repository for runtime, permission, and integration setup instructions.
The description indicates a local-first Chromium automation setup, so an operational Chromium environment and MCP-agent integration are expected. Exact dependencies and version requirements are in the source repository.
Lets MCP-enabled AI read local browser screenshots for analysis and troubleshooting.
Capture responsive website screenshots across viewports and detect visual accessibility issues.
Automate browser tasks, inspect pages, run scripts, and capture diagnostics.
Automate browser tasks, capture console logs, and take screenshots for web workflows.
Drive real local browsers across multiple sessions without cloud or API keys.
Connect AI to a local browser for automated web actions and screenshots.