compare_capabilitiesCompare 2-5 live capabilities using the same real metadata returned by inspect_capability; no fabricated composite score is produced.
| Name | Required | Description | Default |
|---|---|---|---|
| probe | No | ||
| sml_ids | Yes |
Changes observed during successful MCP inspections. Dates show when Glama detected each change.
Input schema / additionalPropertiesAdded value: +falseInput schema / properties / probe / defaultAdded value: +trueInput schema / properties / probe / descriptionRemoved value: -"Attempt real liveness probes when safe to do so (default true)"Input schema / properties / sml_ids / descriptionRemoved value: -"2-5 sml_ids from mesh_discover results to compare"Input schema / properties / sml_ids / items / maxLengthAdded value: +500Input schema / properties / sml_ids / items / minLengthAdded value: +1Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint, openWorldHint, idempotentHint, and destructiveHint, so the bar is lower. The description adds useful behavioral context beyond annotations: it emphasizes that results come from real live metadata and that no fabricated composite score will be generated. This clarifies what the tool will and will not report.
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?
The description is one tightly packed sentence that front-loads the action and scope, then adds a meaningful contrast. No filler or redundancy exists; every clause contributes to selection and invocation understanding.
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?
The core purpose and key boundary ('no composite score') are clear, but with no output schema the description should better indicate what the returned comparison looks like. It also omits any detail about the probe parameter. An agent can select the tool confidently, but may still be uncertain about invocation behavior and output shape.
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 description coverage is 0%, so the description must compensate for parameter meaning. It only weakly implies sml_ids via '2-5 capabilities,' but does not explain that sml_ids are the identifiers being compared. The 'probe' boolean parameter, which defaults to true, is completely unexplained, leaving its behavioral impact ambiguous.
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 names a specific verb ('compare'), a bounded resource ('2-5 live capabilities'), and a concrete data source ('real metadata returned by inspect_capability'). It also distinguishes itself from score-producing tools with 'no fabricated composite score is produced.' This clearly differentiates it from siblings like inspect_capability and recommend_capability.
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 gives clear context: use this when you need a multi-capability comparison grounded in the same real metadata as inspect_capability. The negative clause 'no fabricated composite score is produced' implicitly steers away from recommendation-style tools. However, it does not explicitly name when-not-to-use scenarios or alternative sibling tools.
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.
Most tools have clear boundaries: discover searches, inspect examines one, compare contrasts, recommend interprets natural language, and call executes. Some ambiguity exists between capability_manifest and discover_capabilities for inventory listing, and recommend_capability could be confused with discover_capabilities, but descriptions generally prevent misselection.
All names are lowercase snake_case, and the core capability operations follow a verb_noun pattern. The noun-led capability_manifest, system_status, and x402_* resources deviate from that pattern, and capability names mix singular and plural, but the naming remains readable and internally consistent per subdomain.
Ten tools is well within the ideal range and each tool maps to a distinct part of the capability discovery, comparison, payment, and execution workflow. There are no obviously redundant or throwaway tools, so the count feels appropriately scoped.
The set covers the full read-only market-intelligence lifecycle: inventory, search, inspect, compare, recommend, call, payment readiness, contract lookup, and settlement evidence. Operator CRUD is explicitly outside scope, so there are no meaningful dead ends for an agent using this surface.