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

The description goes well beyond the annotations by disclosing deterministic execution, transient input handling with no storage/logging/retention, compute routing behavior for server vs browser, and the export of an AP2 artifact with execution_hash for chain provenance. It also clarifies gpu:true always delegates to the browser, which is a behavioral trait not inferable from the schema. These are meaningful, non-obvious disclosures that fully align with the readOnly/idempotent annotations.

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 a dense, single-paragraph wall of text that repeats 'OpenChainGraph compute node' twice and mixes core behavioral facts with a URL, FV-status receipt details, and compute-mode explanations. While all information is relevant, it is not structured or prioritized for quick scanning, and every sentence does not earn its place due to redundancy. Conciseness is sacrificed for completeness, making it harder for an agent to extract the essential invocation facts.

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 the tool has no output schema and four parameters including a nested policy_parameters object, the description should clarify what the tool returns and what inputs are needed for the decision function, but it only says an AP2 artifact with execution_hash is exported. The policy_parameters fields are deferred to an external manifest, which an agent may not have access to. The compute-routing and privacy details are thorough, but the missing output semantics and undefined policy parameters leave a significant invitation gap.

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 100%, so the baseline is 3. The description adds a bit of nuance to the compute parameter by explaining the runtime implications of 'auto', 'server', and 'browser', but it does not enrich the meaning of parent_hashes, parent_tool_ids, or policy_parameters beyond what the schema already states. The policy_parameters object is left deferred to 'the tool's manifest', which is not present, so the description does not compensate for the nested object's internal field documentation gap.

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 identifies the tool as a 'Consultation Response Tracker' and an OpenChainGraph compute node, suggesting it computes some kind of analytics artifact, but it never states in concrete terms what the tracker does with consultation inputs or what the 'response' represents. The title is a noun phrase and the name contains 'compute', but the core domain function is left vague. It does distinguish itself from siblings via the 'analytics_mandate' and OpenChainGraph identity, but more as a category than a concrete function.

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 explains compute mode selection and instructs users to 'use synthetic or anonymised inputs only', which is useful invocation guidance, but it does not say when to prefer this tool over any alternative or name sibling tools. There are no exclusions or contextual cues about the typical use case or prerequisites for a consultation response tracking task. The guidance is about how to run the tool, not when to choose it.

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