compare_modelsSide-by-side comparison of pricing ($/1M tokens), context window, and modality for 2-5 OpenRouter models. Answers "is claude-sonnet-4.5 cheaper than gpt-4o". Keyless.
| Name | Required | Description | Default |
|---|---|---|---|
| ids | Yes | Array of 2-5 exact model ids, e.g. ["anthropic/claude-sonnet-4.5", "openai/gpt-4o"]. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations (readOnlyHint, idempotentHint, destructiveHint) already indicate safety. The description adds 'Keyless', implying no authentication needed, and details the comparison dimensions (pricing, context window, modality), which are not in annotations. No contradictions.
Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.
Is the description appropriately sized, front-loaded, and free of redundancy?
Two sentences, no filler. All information is front-loaded and essential: comparison scope, what is compared, an example, and a note on authentication.
Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.
Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?
Given no output schema, the description hints at the output (side-by-side comparison of three attributes). It covers intended use well but lacks explicit return format details.
Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.
Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?
Schema coverage is 100% with a clear description of 'ids' as an array of exact model ids. The description reinforces the 2-5 constraint and provides examples, but does not add new meaning beyond the schema.
Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.
Does the description clearly state what the tool does and how it differs from similar tools?
The description uses a specific verb ('Side-by-side comparison'), identifies the resources (pricing, context window, modality), and constrains scope (2-5 OpenRouter models). The example question clarifies intent and distinguishes from siblings like list_models and get_model.
Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.
Does the description explain when to use this tool, when not to, or what alternatives exist?
The description implies use for comparing model pricing, context window, and modality, exemplified by the question. It does not explicitly state when not to use or name alternatives, but sibling names provide context.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
Several tools overlap in purpose: ask_pipeworx and ask_pipeworx_beta are explicitly identical today, and the polymarket_edges, polymarket_arbitrage, bet_research, and polymarket_fill_risk tools all orbit market-opportunity analysis from slightly different angles. The descriptions are unusually detailed and often say when to prefer one tool over another, but an agent must read carefully to avoid misselection.
Names are consistently snake_case and mostly follow a verb_noun pattern (list_models, get_model, resolve_entity, validate_claim, scan_dependency), with predictable domain prefixes like polymarket_* and pipeworx_*. A few noun-first names like entity_profile, bet_research, and ai_visibility_check deviate slightly, but the overall convention is readable and coherent.
34 tools is well into the 'too many' range, especially for a server named Openrouter that actually spans several unrelated domains: model catalog, Pipeworx data retrieval, Polymarket analysis, memory, subscriptions, and website tooling. Many individual tools are justified, but the set is overstuffed and would be better split into focused servers.
Each sub-domain is reasonably covered: model catalog has list/get/compare, memory has remember/recall/forget, subscriptions have subscribe/list/unsubscribe/recent_alerts, and Pipeworx querying has multiple modes plus discovery. However, a server named Openrouter exposes no way to actually run completions or route requests through OpenRouter, and the unrelated bundled domains make the overall surface feel scattered rather than complete.