Skip to main content
Glama

Pick a provider

decide
Read-onlyIdempotent

First pick, no API key needed, before writing integration code (LLM, sandbox, vector DB, search, email, Postgres, auth, browser, observability). First call: set a stable clientInfo.name (e.g. acme-support-agent), then {task:'run untrusted Python in a sandbox'}. Returns the dated Depot Pick, runner-up, sourced price, when to skip it, reportOutcome (decision id for depot report_outcome), recheck. Recheck: since=recheck.since (unchanged = short reply). Live API error: rescue. Stack changes since last cursor: session_brief. For quotes, compare, audit, freshness or to report an outcome: depot.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "decide"
  }
}

Input Schema

NameRequiredDescriptionDefault
fullNo
jtbdNo
taskNoWhat you need, in words
sinceNoLast recheck.since
existingNo
maxMonthlyNo
constraintsNoe.g. eu-region

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed4 schema fields changed
    • changedInput schema / properties / constraints / description
      Previous value: -"Comma list, e.g. eu-region; may switch the pick"New value: +"e.g. eu-region"
    • removedInput schema / properties / existing / description
      Removed value: -"Provider ids you already use"
    • removedInput schema / properties / jtbd / description
      Removed value: -"Or a job id, e.g. agent-sandbox"
    • removedInput schema / properties / maxMonthly / description
      Removed value: -"USD budget for autoActOk"
  2. Changed3 schema fields changed
    • changedInput schema / properties / existing / description
      Previous value: -"Comma list of provider ids you already use"New value: +"Provider ids you already use"
    • changedInput schema / properties / jtbd / description
      Previous value: -"Or a job id, e.g. agent-sandbox; wins over task"New value: +"Or a job id, e.g. agent-sandbox"
    • addedInput schema / properties / since
      Added value: +{
      +  "description": "Last recheck.since",
      +  "type": "string"
      +}
  3. Changed3 schema fields changed
    • changedInput schema / properties / constraints / description
      Previous value: -"Comma list checked against the skip rule"New value: +"Comma list, e.g. eu-region; may switch the pick"
    • changedInput schema / properties / existing / description
      Previous value: -"Provider ids you already use"New value: +"Comma list of provider ids you already use"
    • changedInput schema / properties / jtbd / description
      Previous value: -"Or a job id, e.g. coding-agent"New value: +"Or a job id, e.g. agent-sandbox; wins over task"
  4. Changed7 schema fields changed
    • changedInput schema / properties / constraints / description
      Previous value: -"comma list, optional"New value: +"Comma list checked against the skip rule"
    • addedInput schema / properties / existing
      Added value: +{
      +  "description": "Provider ids you already use",
      +  "type": "string"
      +}
    • removedInput schema / properties / full / description
      Removed value: -"Return full plan tables and runtime route arrays (default false, compact < 4 KB)"
    • changedInput schema / properties / jtbd / description
      Previous value: -"JTBD id, e.g. coding-agent"New value: +"Or a job id, e.g. coding-agent"
    • addedInput schema / properties / maxMonthly
      Added value: +{
      +  "description": "USD budget for autoActOk",
      +  "minimum": 0,
      +  "type": "number"
      +}
    • addedInput schema / properties / task
      Added value: +{
      +  "description": "What you need, in words",
      +  "type": "string"
      +}
    • removedInput schema / required
      Removed value: -[
      -  "jtbd"
      -]
  5. First observed

TDQS

A4/5.0
Behavior4/5

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

Annotations already cover the safety profile (readOnly, idempotent, non-destructive), but the description adds real context beyond them: it returns a dated pick plus runner-up, sourced price, a skip condition, a reportOutcome decision id, and a recheck cursor, and it notes no API key is required. It stops short of stating rate limits or what a 'short reply' on unchanged recheck actually contains.

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?

Front-loaded and information-dense with zero filler, but written in a telegraphic, parenthetical style that compresses three distinct concerns (first call, return payload, sibling routing) into one block. The routing sentences are efficient; the return-payload sentence is a comma-spliced list that is harder to parse than it needs to be.

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 7-parameter read-only tool with no output schema, the description is unusually complete: it discloses what comes back (pick, runner-up, price, skip condition, decision id, recheck) and how to refresh. It leaves gaps on four optional parameters and on the jargon of 'Depot Pick', but nothing needed to invoke it correctly is missing.

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?

Coverage is 43% (below half), so the description needs to compensate and only partly does: it exemplifies 'task' ({task:'run untrusted Python in a sandbox'}) and fully explains 'since' via the recheck cursor. Four schema parameters (full, jtbd, existing, maxMonthly) get no explanation, and the description introduces 'clientInfo.name' which is not an input-schema property at all, leaving the caller to map that onto the actual schema.

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?

States a specific verb and resource ('First pick' of a provider to integrate) and enumerates the domains it covers (LLM, sandbox, vector DB, search, email, Postgres, auth, browser, observability), so an agent can tell it is a recommendation tool rather than a lookup or quote tool. Sibling differentiation is explicit ('depot' for quotes/compare/audit, 'rescue' for API errors). It loses a point for jargon like 'Depot Pick' and 'recheck' that are undefined on first read.

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?

Gives an explicit sequencing rule ('First pick ... before writing integration code'), explicit recheck semantics ('since=recheck.since (unchanged = short reply)'), and routes every adjacent need to a named alternative: depot for quotes/compare/audit/freshness/report outcome, rescue for live API errors, session_brief for stack changes. When-to-use and when-to-use-something-else are both covered.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources