Skip to main content
Glama

Compass by 402rates

by com.402rates
OverviewDiscussionsAdmin

Start or repeat a synthetic agent test

start_agent_test

Create a free bounded synthetic run without an account. An optional finalized previous run and its token link a repeat of the same case version and scenario. No real purchase. Caller model costs remain external.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
case_idYesVersioned synthetic case identifier.
variantNoBudget case: budget_first or attempts_first. Linked repeats retain the previous variant.
access_tokenNoSecret short-lived run token. Omitted from reports. MCP client history may retain this argument.
case_versionNoOptional exact case version. A new run defaults to the current published version; a linked repeat retains its previous version. The resolved version is returned and stored.
previous_run_idNoOptional finalized run to repeat. Its access token authenticates this start.
configuration_changedNoOptional caller-declared change since the previous run; not independently verified.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed2 schema fields changed
    • changedInput schema / properties / case_version / description
      Previous value: -"Exact case version. No implicit migration."New value: +"Optional exact case version. A new run defaults to the current published version; a linked repeat retains its previous version. The resolved version is returned and stored."
    • addedInput schema / required
      Added value: +[
+  "case_id"
+]
  2. Changed1 schema field changed
    • addedInput schema / properties / variant
      Added value: +{
+  "description": "Budget case: budget_first or attempts_first. Linked repeats retain the previous variant.",
+  "maxLength": 32,
+  "type": "string"
+}
  3. First observed

TDQS

A4.2/5.0
Behavior4/5

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

With only minimal annotations, the description carries the behavioral burden and adds valuable traits: the run is free, bounded, account-free, non-purchase, and caller model costs remain external. It does not contradict the readOnlyHint=false/destructiveHint=false annotations, and it gives more context than the annotations alone.

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 four short sentences with no filler, and the core purpose is front-loaded in the first sentence. Economic and repeat-mode details each earn their place without bloating the text.

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?

For a six-parameter tool without an output schema, the description covers the main preconditions (no account), side effects/costs (free, bounded, no purchase, external model costs), and repeat semantics. The parameter schema fills the remaining invocation details, though the exact return payload is left unspecified.

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

Parameters3/5

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

Schema description coverage is 100%, so the structured fields already document case_id, variant, access_token, case_version, previous_run_id, and configuration_changed. The description adds the relationship that a finalized previous run plus its token triggers a linked repeat, but it does not need to compensate for schema gaps.

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 uses a specific verb+resource ('Create a free bounded synthetic run') and explains the repeat mode ('An optional finalized previous run and its token link a repeat of the same case version and scenario'). This clearly separates start_agent_test from sibling tools that describe, inspect, list, or read test artifacts.

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

Usage Guidelines4/5

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

It gives clear context for when to use the tool: creating or repeating a synthetic run without an account, and it explicitly signals that no real purchase is involved, which avoids confusion with payment-related siblings. It does not name explicit alternatives or when-not-to-use conditions, so it stops short of a 5.

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

All Compass by 402rates toolsInspect
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