prepare_observations

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

The annotations already carry readOnlyHint=true, idempotentHint=true, destructiveHint=false and openWorldHint=false, so the safety profile is covered without the description. What the description adds beyond the structured fields is genuinely useful and non-obvious: that no LLM provider is invoked and the agent must supply its own inference. It still omits pagination/limit behavior and whether the result is cached, so it is not a 5.

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 short sentences, no filler, and the most decision-relevant fact ('Use your own model; no provider call') is placed where it will be read. The first sentence is dense with undefined jargon, which costs a little clarity but not length.

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?

There is no output schema to lean on, 0% schema description coverage on three parameters, and no indication of what 'prepared' material looks like or how the agent should consume it. For a preparatory read tool whose whole value is the shape of what it hands back, this leaves the agent under-equipped.

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?

There are three parameters (limit, source_id, start_byte) at 0% schema description coverage, and the description says nothing about any of them. The names are partly self-explaining (source_id, start_byte as a byte offset), but no defaults, units, or interaction (e.g., what limit bounds, whether start_byte pairs with source_id) are given, so the agent must guess.

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 verb 'prepare' and the three artifacts (source excerpts, claim targets, interpretation contract) are stated, so the tool is not a pure tautology of its name. However, the artifacts are undefined domain jargon and the description gives no differentiation from the many sibling tools (e.g., prepare_memory_refresh), so an agent cannot tell exactly what is produced. This is a vague but non-empty purpose.

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 only usage signal is 'Use your own model; no provider call,' which hints at the calling pattern but never states when this tool should be selected over siblings such as prepare_memory_refresh, recall_project, or read_project_source. No preconditions, no exclusions, no alternative routing.

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