MCP tool surface

15 tools. Plain English in, structured results out.

alethia_tell

Execute natural-language E2E test instructions against the page Alethia is currently driving. Returns per-step results, policy audit records, and a SHA-256 integrity hash. Destructive actions (delete, purchase, transfer, etc.) are blocked unconditionally by the EA1 gate. ~13 ms per step on average.

ParamTypeRequiredDescription
instructionsstringyesOne or more plain-English test instructions, newline-separated.
namestringnoShort label for the run, shown in cockpit history and audit trails. Strongly recommended.
allowSensitiveInputbooleannoAllows typing into password, token, credit card, and other sensitive fields. Only for legitimate auth/payment flow tests.
alethia_tell({
  name: "signin smoke",
  instructions: "navigate to http://localhost:3000\nclick Sign In\nassert the dashboard is visible"
})

alethia_tell_parallel

Run multiple test flows concurrently, each against a different URL. Spawns a browser instance per spec and returns all results together. Use it to verify multiple pages at once.

ParamTypeRequiredDescription
specsarrayyesEach item has url, instructions, and an optional name.
alethia_tell_parallel({
  specs: [
    { url: "http://localhost:3000/", instructions: "assert Dashboard is visible" },
    { url: "http://localhost:3000/settings", instructions: "assert Settings is visible" }
  ]
})

alethia_compile

Compile natural-language test instructions to Alethia's Action IR without executing anything. Returns the compiled IR, per-line confidence scores, and warnings for lines the compiler couldn't parse. Use it to preview a tell() call, debug coverage gaps, or generate reproducible IR scripts for CI.

ParamTypeRequiredDescription
instructionsstringyesPlain-English instructions to compile.
alethia_compile({ instructions: "click Sign In\nassert Dashboard is visible" })

alethia_eval

Evaluate a JavaScript expression in the page under test and return the result. Runs in the context of the navigated page, not the host UI. Use it for queries the NLP compiler can't express — counting elements, reading computed styles, checking localStorage.

ParamTypeRequiredDescription
expressionstringyesJavaScript expression to evaluate in the page context.
alethia_eval({ expression: "document.querySelectorAll('li').length" })

alethia_screenshot

Capture a PNG screenshot of the current page and return it as a base64-encoded image. Use this to visually verify what the browser is showing after running tell() steps.

No parameters.

alethia_screenshot()

alethia_status

Health and identity probe. Returns runtime version, the active VITRON-EA1 policy profile, kill switch state, driver statistics, current page domain, and runtime capabilities. Call this for liveness checks before sending tell() calls.

No parameters.

alethia_status()

alethia_assert_safety

Navigate to a URL, discover every destructive / write-high action on the page, and verify the VITRON-EA1 policy gate blocks each one. Returns a per-action block/allow report — the automated way to prove the safety gate holds on a real page before releasing an agent-driven workflow against a customer environment.

ParamTypeRequiredDescription
urlstringyesThe URL to navigate to and audit.
alethia_assert_safety({ url: "http://localhost:3000/admin" })

alethia_activate_kill_switch

Halt all current and queued automation immediately. The EA1 policy gate stays armed; subsequent tell() calls are blocked with reason KILL_SWITCH_ACTIVE until reset. The kill switch auto-clears on the operator's next Run from the cockpit — agents have no self-release path, by design.

ParamTypeRequiredDescription
reasonstringnoOptional explanation that lands in the audit trail.
alethia_activate_kill_switch({ reason: "unexpected navigation off localhost" })

alethia_audit_wcag

Run a WCAG 2.1 AA accessibility scan on the current page — 14 DOM-level checks including missing alt text, unlabeled form fields, missing lang attribute, generic link text, empty headings, and duplicate IDs. Call after navigating with alethia_tell.

This is a fast heuristic scan for catching common issues early — it is not a substitute for a full axe-core-level audit or a manual accessibility review, and does not by itself constitute a WCAG conformance claim.

Free to evaluate. Production and commercial use requires a commercial license — email team@vitron.ai.

No parameters.

alethia_audit_wcag()

alethia_audit_nist

Run a lightweight security-hygiene scan on the current page, checking 8 client-visible proxies loosely mapped to NIST SP 800-53 control families — AC (login lockout, session timeout), IA (unmasked password fields, MFA-related markup), and SI (input validation, error-message leakage). Call after navigating with alethia_tell.

Most NIST SP 800-53 controls are organizational or infrastructure controls that can't be assessed by inspecting a rendered page. Treat this as an early signal during development, not as compliance evidence for an audit or ATO package.

Free to evaluate. Production and commercial use requires a commercial license — email team@vitron.ai.

No parameters.

alethia_audit_nist()

alethia_export_session

Export the full session recording as a signed evidence pack: every tool call, input, output, and policy decision, plus a SHA-256 integrity hash. Use it at the end of an agent loop to produce cryptographic proof of everything the agent did — built for compliance review and chain-of-custody.

Free to evaluate. Production and commercial use requires a commercial license — email team@vitron.ai.

No parameters.

alethia_export_session()

alethia_propose_tests

Navigate to a URL, scan it for interactive elements, and generate a candidate NLP test suite ready to pass to alethia_tell — including an auto-generated "EA1 Safety Gate Verification" block with an expect block: line for every destructive control found. Use it to bootstrap coverage on a page you haven't tested yet.

ParamTypeRequiredDescription
urlstringyesThe URL to navigate to and scan.
alethia_propose_tests({ url: "http://localhost:3000/" })

alethia_show_cockpit

Show the Alethia cockpit window — the oversight surface where the target app is driven and each step is highlighted live (green = pass, blue = type, red = EA1 block). Use it to pop the UI into view during a headless-launched session for demos or review.

No parameters.

alethia_show_cockpit()

alethia_hide_cockpit

Hide the Alethia cockpit window. The runtime keeps running and continues to accept tool calls — only the visible window is dismissed.

No parameters.

alethia_hide_cockpit()