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

Annotations already cover read-only/idempotent/non-destructive/non-open-world, but the description adds substantial behavior: bounded chunking, offset+sha256 continuation semantics, the caveat that a check's canonical digest differs from the serialization digest, and the hard limits (no recovery of removed content, no target invocation). These are non-obvious operational traits not inferable from the 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?

Front-loaded with the read scope, then pagination, then limits; each sentence carries distinct information. It is dense and jargon-heavy ('not an omitted check's different canonical digest'), but for a 10-parameter tool the overall length is disciplined.

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?

For a complex tool with no output schema and 0% schema descriptions, the description does mention returned signals (nextOffset, serialization sha256), which helps. It still does not describe the shape/nature of the returned chunk, the meaning of 'bounded', or how the nine source modes differ in practice, leaving real gaps for an agent invoking it blind.

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 0%, so the description carries the full burden for 10 parameters. It meaningfully explains the interdependent ones (offset first-omit then nextOffset, sha256 as exact serialization digest, pointer as JSON Pointer field selection, checkId pairing with check_actual/check_expected), but leaves limit bounds, projectId/environmentId/runId roles, and the source enum's practical distinctions to inference.

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?

States a specific verb ('read') and a precise, enumerated resource set: bounded chunks of retained run output/input/scenario/configuration/target/assessment, a cited event, or check_actual/check_expected operands. An agent can tell this is a targeted evidence-extraction tool rather than a lister. It stops short of naming which sibling (e.g., firedrill_event, firedrill_artifacts) it displaces.

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?

Gives procedural guidance ('omit offset initially; continue with returned nextOffset and sha256') and a when-not ('cannot recover missing/removed content or invoke a target'), which is real exclusion context. However, across 45 siblings it never routes the agent to an alternative for the cases where this tool is the wrong choice.

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