record_project_science_evidence

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

The description richly exceeds the annotations: it declares the call 'records context but authorizes no training, deployment, policy activation, skill update, or production action,' which is precisely the side-effect boundary an agent needs to avoid over-triggering downstream work. It also discloses the confirm_record inspection requirement and append semantics consistent with idempotentHint=false. No contradiction with readOnlyHint=false or destructiveHint=false.

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?

Four sentences (~90 words), each with a distinct job: summary of the core action, trigger timing, the confirm gate, and the no-side-effect boundary. The main action is front-loaded and there is no repetition or filler.

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?

Covers the three highest-risk aspects — when to call, the confirm_record gate, and what the call does NOT authorize — and the presence of an output schema relieves it of explaining return values. But with 10 required parameters and zero schema descriptions, the definition is thin on parameter-level semantics; an agent must infer formats for the JSON payload fields and artifact identifiers.

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 carries the burden. It explains confirm_record ('only after inspecting'), implicitly conditions parent_receipt_ids_json ('Later stages must cite predecessor receipt IDs'), and enumerates artifact_kind values. However, 16 of 19 parameters (evidence_id, observed_at, the *_json payloads, artifact_system, artifact_digest_sha256, business_metric_id, etc.) receive no semantic help, leaving significant gaps for a call with 10 required fields.

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 ('Append'), a resource ('bounded scientific-lineage receipt'), and a condition ('after explicit review'). The second sentence scopes the domain by naming artifact types (hypothesis, search, data, AutoML, solver), which differentiates it from sibling record_* tools like record_project_learning_review. Loses a point because 'scientific-lineage receipt' is jargon-heavy and no sibling is explicitly named for contrast.

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?

'Use this immediately after ... artifact is durably recorded' gives an explicit trigger condition, and 'Later stages must cite predecessor receipt IDs' explains its role in a workflow chain. The confirm_record gate ('only after inspecting the artifact and its source event') adds a precise when-to-invoke qualifier. No alternative tools are named or excluded, so it stops short of full routing guidance.

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