actionlayer

actionlayer_start_task

Execute a real-world goal on the web — book, order, fill, click, log in, extract, automate. Use this when the user wants something DONE on a live website (Resy, Amazon, any URL), not when they want analysis or text.

An AI agent runs the task end-to-end on the live web — it reads the page and adapts to whatever it finds (CAPTCHA, novel UI, login wall, weird checkout). Tickets that would normally fail on a hard step just keep progressing. You don't need to design around the unhappy path.

Prefer actionlayer_invoke_action with a typed action id when one fits — it's faster and more deterministic. Reach for actionlayer_start_task when no typed action covers the site, or when the goal mixes multiple steps.

Be SPECIFIC. The goal must include: which site/place, when, how many, exact product URL if buying. Vague goals ("book me dinner", "order something") block on the user and waste planning budget — ask the user for the missing pieces FIRST, then call this with everything nailed down.

Input Schema

TableJSON Schema

Name	Required	Description
`goal`	Yes	Specific natural-language description. Good: "Book a table for 2 at Bar Crenn (SF) on 2026-12-06, 7-8:30pm via Resy" Bad: "find me a flight" (too vague — will block)
`webhook_url`	No	Optional. Best-effort POST (single attempt, no retries) when the ticket blocks on user input or reaches a terminal state. Thin payload {kind, ticket_id, state, occurred_at} — treat it as a poke and call actionlayer_get_task for the details. Polling works with or without it.
`max_budget_usd`	No	Required for anything that spends money. The planner soft-caps spend and escalates instead of overspending.
`idempotency_key`	No	Optional caller-supplied retry key — same key within 24h returns the same ticket.

Output Schema

TableJSON Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

First observedJul 14, 2026

TDQS

A4.7/5.0

Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations provided, so description carries full burden. It discloses that an AI agent runs the task end-to-end, adapts to CAPTCHA/login walls, and that tickets progress on hard steps. Mentions webhook behavior and budget soft-cap. However, it could be more explicit about idempotency handling and potential side effects.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is well-structured and front-loaded. First paragraph states purpose, second gives behavioral context and preference for alternative, third provides critical usage advice. Every sentence adds value with no wasted words.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the complexity, parameter count, and presence of output schema, the description covers all essential aspects: purpose, usage, behavioral traits, parameter advice, and differentiation from siblings. It does not need to detail return values as output schema exists.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is 100%, so parameters are fully described. The description adds valuable guidance on goal specificity and examples, reinforcing the schema. The webhook parameter behavior is well explained in the description.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states that the tool executes real-world goals on the web (book, order, fill, click, log in, extract, automate). It also distinguishes itself from the sibling tool actionlayer_invoke_action by specifying when to prefer that alternative.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Provides explicit when-to-use (user wants something done on a live website) and when-not-to-use (analysis or text). Explicitly names the alternative actionlayer_invoke_action with conditions for preference. Also advises on being specific and asking for missing details.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All actionlayer tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?