Lastest MCP server

Read from the package source without running it. The installed server may list more.

• lastest_approve_all_diffs

  Approve all pending visual diffs in a build at once.

• lastest_approve_baseline

  Approve visual changes, updating baselines. Accepts one or more diff IDs.

• lastest_approve_diff

  Approve a single visual diff, accepting the current screenshot as the new baseline.

• lastest_approve_layer

  Per-layer feedback on a step comparison: approve (Mark expected \u2192 write baseline), reject (Needs fix \u2192 create todo), or snooze (suppress for this build only).

• lastest_create_areawrite action

  Create a new functional area for organizing tests.

• lastest_create_repowrite action

  Create a new local repository in the current team. Use this when you need a fresh workspace for tests without connecting to GitHub/GitLab. Pass `baseUrl` to point the repo at an external app (e.g. https://staging.example.com) so generated tests target the right origin instead of localhost:3000.

• lastest_create_setup_scriptwrite action

  Create a reusable setup script for a repository. Two types: `playwright` (async function with page+context) and `api` (HTTP seeding). Attach the returned ID to a test via lastest_update_test with `setupScriptId`.

• lastest_create_storage_statewrite action

  Create a new storage state for a repo from a Playwright `storageState()` JSON string. The JSON should be the output of `await context.storageState()` \u2014 an object with `cookies` and `origins` arrays. Use this to share a logged-in session across tests instead of re-logging in. Treat the raw JSON like a password: only call this with tokens you intend Lastest to use.

• lastest_create_testwrite action

  Create a test. Two modes: (1) **direct** \u2014 pass { name, code } to insert ready-made Playwright code. (2) **AI** \u2014 pass { url } and/or { prompt } to have the Lastest AI agent generate the test. Direct mode returns immediately with a test ID; AI mode may take longer and returns the generated code.

• lastest_delete_areawrite action

  Soft-delete a functional area. Tests remain but become unassigned.

• lastest_delete_setup_scriptwrite action

  Delete a setup script. Refused with 409 if any test still references it via setupScriptId \u2014 the response lists the blocking tests.

• lastest_delete_storage_statewrite action

  Delete a storage state. Tests referencing it via setupOverrides.extraSteps will lose that step \u2014 review usage before deleting.

• lastest_delete_testwrite action

  Soft-delete a test (can be restored). Does not permanently remove test data.

• lastest_get_build_status

  Get the current status and results of a build. Use after lastest_run_tests to check progress. Returns slim build scalars + slim diff index \u2014 drill into a specific diff with lastest_get_visual_diff (or pass `includeDiffs: "full"` if you really need the joined a11y/network/AI payloads).

• lastest_get_change_map

  Get the build-level Change Map (4-signal area ranking + AI intent/risk summary) for a build.

• lastest_get_coverage

  Get test coverage statistics by functional area and route.

• lastest_get_diff

  Get full details of a single visual diff including pixel data, AI analysis, and test info.

• lastest_get_job_status

  Get the status and progress of a specific background job.

• lastest_get_playwright_settings

  Get the repo-level Playwright settings (browser, viewport, timeouts, error modes, stabilization, parallelism caps, etc.). Falls back to the global row + built-in defaults when no per-repo row exists.

• lastest_get_repo

  Get details about a specific repository.

• lastest_get_setup_script

  Get a setup script including its code.

• lastest_get_test

  Get full details of a single test including code, URL, and last run status.

• lastest_get_test_runwrite action

  Get detailed results for a specific test run, including individual test results, errors, and durations.

• lastest_get_visual_diff

  Get visual diff details for a build, including AI classification and confidence scores.

• lastest_heal_test

  Trigger the AI healer agent to automatically fix a failing test by inspecting the live UI and updating selectors/assertions.

• lastest_health_check

  Check connectivity to the Lastest instance.

• lastest_list_active_jobs

  List currently active background jobs (builds, AI operations, etc.).

• lastest_list_areas

  List functional areas (test groupings) for a repository.

• lastest_list_build_shares

  List public shares anchored on a build (includes revoked ones \u2014 check `status`). Use to find an existing share before publishing a duplicate, or to grab a slug for revoke.

• lastest_list_builds

  List recent builds for a repository with status and test counts.

• lastest_list_failing_tests

  List tests that are currently failing, with error details.

• lastest_list_repos

  List all repositories accessible to the current team.

• lastest_list_setup_scripts

  List setup scripts for a repository. Each entry includes id, name, type (`playwright` or `api`), and code so you can pick one to attach to a test via setupScriptId.

• lastest_list_storage_states

  List storage states (saved Playwright `storageState()` blobs \u2014 cookies + localStorage) for a repository. Returns metadata only; the raw JSON is omitted because it contains live auth tokens. Use these IDs when wiring `setupOverrides.extraSteps` on a test via lastest_update_test.

• lastest_list_test_shares

  List public shares anchored on a single test.

• lastest_list_tests

  List all tests in a repository with their latest pass/fail status.

• lastest_list_tests_by_area

  List tests within a specific functional area.

• lastest_publish_sharewrite action

  Publish a public-share link for a build (or a single test within it). Returns a `/r/<slug>` URL anyone can view without logging in. Use after a build completes so demos and outreach messages can link directly to the visual result. Pass `scopedTestId` to scope the share to one test instead of the whole build.

• lastest_qa_summary

  Get a comprehensive QA overview for a repository: test health, recent builds, and action items.

• lastest_reject_baseline

  Reject visual changes, marking them as regressions. Accepts one or more diff IDs.

• lastest_reject_diff

  Reject a single visual diff, marking it as a regression.

• lastest_review_build

  Comprehensive build review: fetches build details, visual diffs, and failed tests into a structured QA summary with action items.

• lastest_revoke_share

  Revoke a public share by its share ID. The `/r/<slug>` URL stops resolving immediately. Find share IDs via lastest_list_build_shares or lastest_list_test_shares.

• lastest_run_testswrite action

  Trigger a test build and return structured results. Can run all tests in a repo, all tests inside one functional area, or specific test IDs. Pass `forceVideoRecording: true` when you intend to publish a share that should include the video player (the share page renders the embedded clip only when the build has video).

• lastest_update_areawrite action

  Update a functional area's name, description, or parent.

• lastest_update_playwright_settingswrite action

  Upsert repo-level Playwright settings. Pass any subset of fields \u2014 unknown keys are rejected, unspecified ones stay at their current value. Use this when the demo skill needs to retarget a repo at a different browser, viewport, or error policy without touching every test individually.

• lastest_update_repowrite action

  Update a repository's name, default branch, selected branch, or base URL. `baseUrl` is the value passed to `test(page, baseUrl, ...)` \u2014 set it to point a repo at an external app instead of the default http://localhost:3000.

• lastest_update_setup_scriptwrite action

  Update a setup script's name, type, code, or description.

• lastest_update_testwrite action

  Update a test's name, code, URL, functional area, setup wiring, or runtime overrides. Pass `setupTestId` to point this test at another test for setup (the most common pattern when two tests share a login), or `setupScriptId` to point at a saved setup script \u2014 these two are mutually exclusive. Use `setupOverrides` / `teardownOverrides` to skip specific default steps or inject extra ones (test

• lastest_verify_build

  Get the full verify-build view: Change Map + step comparisons grouped by regression vs intent gate.

• Code scan5 source files scanned25/25
• –Live reliabilityno gateway calls yet and no remote to proben/a
• –Tool poisoningtools not inspected (local package is not executed); not countedn/a
• Auth qualitylocal package, no credentials required12/15
• Maintenancelast push 7 days ago15/15
• Maintainer identityregistry namespace matches repository owner7/10

Safety score

91/100

higher than 85% of 48,255 servers

Tools

50

the typical server has 7

Tools that change things

34%

typical: 0%. Fewer write tools means less an agent can break

GitHub stars

19

more than 91% of 28,688 servers

Downloads a week

40

more than 31% of 12,191 servers with a package

Lastest: common questions

Is Lastest MCP server safe?

Yes, by our scan: it is graded A (91/100). Read the Lastest safety report

How do I install Lastest?

It runs on your machine. Copy the Claude Code, Cursor, VS Code or Claude Desktop config from the install section.

Does Lastest need an API key?

Not as far as the registry entry and our scan can tell: no credentials are declared or required.

Is Lastest maintained?

The last commit was 46 days ago (2026-09-19). The latest release is v0.3.7.

How does Lastest compare with other MCP servers?

safety score 91/100 (higher than 85% of 48,255 servers); tools 50 (the typical server has 7); tools that change things 34% (typical: 0%. Fewer write tools means less an agent can break). Full comparison.