Skip to main content
Glama

Observe Web Execution

observe_web_execution
Destructive

Arms a selected CDP page for a finite coverage window, resetting counters and returning precise UTF-16 execution ranges plus hashed script text and request initiators for externally driven activity.

Instructions

Instrument one selected CDP page for a finite, externally driven precise coverage window. Resets execution counters and prevents optimized execution until stopped. Returns complete nested UTF-16 ranges/counts, independently hashed script text and request initiator associations by session script ID. Does not navigate, click, evaluate selected code or close the externally owned page. Navigation ends document-scoped collection; workers and UI causality remain outside this claim. Emits an actual armed progress notification before external actions.

Example tool call

This call uses example arguments supplied in the tool’s input schema.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "target_id": "TARGET_ID_FROM_LIST_BROWSER_TARGETS",
      "cdp_endpoint": "http://127.0.0.1:9222",
      "observation_ms": 10000
    },
    "name": "observe_web_execution"
  }
}

Input Schema

NameRequiredDescriptionDefault
target_idYesExact CDP target identifier selected from the endpoint's target listing.
cdp_endpointYesLiteral loopback Chrome DevTools Protocol endpoint.
observation_msNo
allowed_originsNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

B3.4/5.0
Behavior4/5

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

The description explains the mutation that annotations already classify as destructive: it resets execution counters and suppresses optimized execution until the observation is stopped. That explains why the tool is both destructive and non-idempotent. It also states that a page owned externally is not closed and that navigation truncates collection, which is meaningful operational context. It adds little about error handling or what the 'armed progress notification' actually looks like.

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

Conciseness3/5

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

It is front-loaded with the core action and compact, but the writing is heavily compressed, and terms such as 'externally driven', 'causality', and 'armed progress notification' slow the agent down. Most sentences carry information, though the phrasing is closer to internal design notes than agent-facing guidance.

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

Completeness4/5

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

No output schema exists, and the description does describe the return payload: nested UTF-16 ranges and counts, independently hashed script text, and request initiator associations. That is solid coverage for a destructive mutation tool whose annotations already flag its side-effect profile. The unresolved gaps are the allowed_origins parameter and the mechanics of the progress notification.

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

Parameters2/5

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

Schema description coverage is only 50%: observation_ms and allowed_origins have no schema descriptions at all. The description hints at a finite window with 'finite, externally driven precise coverage window', which loosely corresponds to observation_ms, but it says nothing about allowed_origins or its scoping effect. With a half-documented schema, the description should compensate but largely does not.

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

Purpose4/5

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

The description names a specific verb and resource: instrument one selected CDP page to collect a precise coverage window. It also scopes that window to a finite, externally driven session. It is distinguishable from a general session or runtime observer. It stops short of expressly contrasting itself with observe_web_session, so it lacks sibling differentiation of the highest order.

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

Usage Guidelines3/5

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

The exclusions are useful usage boundaries: it does not navigate, click, evaluate selected code or close the page. In addition, navigation ends document-scoped collection, and workers and UI causality are out of scope. However, the definition never says when an agent should choose this tool over observe_web_session or observe_javascript_runtime; the conditions for use must be inferred from the stated limitations.

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

Deploy Server

Other Tools