Skip to main content
Glama

Poly-Glot AI Workspace

by io.github.hmoses
OverviewDiscussionsAdmin

Prepare custom Compare Mode

prepare_custom_compare

Prepare a canonical prompt and instructions for comparing at least two targets, including one to five custom BYOM models and optional supported built-in AI providers. API keys must not appear in the comparison plan; use run_custom_model to execute custom targets separately. Returns comparison targets and prompt without invoking any models. This is a Send action requiring an active Apple-authorized introductory trial, active Pro subscription, or existing unexpired legacy app-managed trial. MCP cannot start an Apple subscription or paid trial.

Input Schema

NameRequiredDescriptionDefault
promptYesRequired canonical prompt shared among all selected models, up to 30,000 characters.
uiLanguageNoInterface language name or code, e.g. EN; independent of source and response languages.EN
customModelsYesRequired array of 1 to 5 custom model connection descriptors; excludes authentication credentials.
inputLanguageNoLanguage name or code of input text, e.g. es; defaults to EN.EN
outputLanguageNoRequested output language name or code, e.g. fr; defaults to EN.EN
builtinProvidersNoOptional built-in provider IDs to include alongside custom targets. This preparation tool does not call providers.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • changedInput schema / properties / builtinProviders / items / enum
      Previous value: -[
-  "chatgpt",
-  "claude",
-  "gemini",
-  "perplexity",
-  "grok",
-  "copilot",
-  "mistral",
-  "huggingchat",
-  "duckduckgo"
-]New value: +[
+  "chatgpt",
+  "claude",
+  "gemini",
+  "perplexity",
+  "grok",
+  "mistral",
+  "huggingchat",
+  "duckduckgo"
+]
  2. Changed12 schema fields changed
    • addedInput schema / properties / builtinProviders / description
      Added value: +"Optional built-in provider IDs to include alongside custom targets. This preparation tool does not call providers."
    • changedInput schema / properties / builtinProviders / items / enum
      Previous value: -[
-  "chatgpt",
-  "claude",
-  "gemini",
-  "perplexity",
-  "grok",
-  "copilot",
-  "mistral"
-]New value: +[
+  "chatgpt",
+  "claude",
+  "gemini",
+  "perplexity",
+  "grok",
+  "copilot",
+  "mistral",
+  "huggingchat",
+  "duckduckgo"
+]
    • addedInput schema / properties / customModels / description
      Added value: +"Required array of 1 to 5 custom model connection descriptors; excludes authentication credentials."
    • addedInput schema / properties / customModels / items / properties / adapterMode / description
      Added value: +"Adapter type for this model's API."
    • addedInput schema / properties / customModels / items / properties / baseUrl / description
      Added value: +"Public HTTPS base URL for an OpenAI-compatible model."
    • addedInput schema / properties / customModels / items / properties / endpoint / description
      Added value: +"Public HTTPS inference URL for a custom REST model."
    • addedInput schema / properties / customModels / items / properties / label / description
      Added value: +"Human-readable name of the custom model in the comparison results."
    • addedInput schema / properties / customModels / items / properties / model / description
      Added value: +"Optional external model identifier."
    • addedInput schema / properties / inputLanguage / description
      Added value: +"Language name or code of input text, e.g. es; defaults to EN."
    • addedInput schema / properties / outputLanguage / description
      Added value: +"Requested output language name or code, e.g. fr; defaults to EN."
    • addedInput schema / properties / prompt / description
      Added value: +"Required canonical prompt shared among all selected models, up to 30,000 characters."
    • addedInput schema / properties / uiLanguage / description
      Added value: +"Interface language name or code, e.g. EN; independent of source and response languages."
  3. First observed

TDQS

A4.8/5.0
Behavior5/5

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

Annotations only say readOnlyHint=false, destructiveHint=false, openWorldHint=false. The description adds substantial context beyond that: it is a 'Send action,' it returns targets and prompt without model invocation, API keys must not appear in the plan, and it carries gating prerequisites. This is the kind of behavioral disclosure a mutation-flagged tool needs.

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

Conciseness4/5

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

Front-loads what the tool produces, then the alternative, then constraints and auth prerequisites. Dense and mostly waste-free, though the subscription/entitlement sentence is long and could be trimmed slightly without losing meaning.

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?

With no output schema, the description still tells the agent what comes back ('returns comparison targets and prompt'), what does not happen (no model invocation), what is required to call it (entitlement), and where execution goes next. Nothing material for correct invocation is missing.

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 coverage is 100%, so the baseline is 3, but the description adds two constraints not obvious from the schema: the 'at least two targets' requirement and the prohibition on API keys appearing in the comparison plan, plus the notion that the prompt is shared canonically across all selected models.

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?

States a specific verb and resource (prepare a canonical prompt + comparison targets) and bounds the scope precisely: 1–5 custom BYOM models plus optional built-in providers. It is immediately distinguishable from the sibling prepare_compare (built-in only) and from run_custom_model (which actually executes).

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?

The description explicitly routes the agent: 'use run_custom_model to execute custom targets separately,' and clarifies this tool does not invoke any models. It also states the exact conditions under which the tool is available (active trial, Pro subscription, unexpired legacy trial) and the limitation that MCP cannot start a subscription or paid trial.

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

All Poly-Glot AI Workspace 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