research_finish

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

With no annotations, the description carries the full burden and does substantial work: it states the operation publishes only when checks pass, that it does not send messages or publish externally, that length_constraint rows are measured but no prose limit is inferred, and that new ordinary-review attempts require schema_version 2 plus decision_actor and hashes. Auth/permission and reversibility are still unstated, keeping it below 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?

The text is a dense cluster of run-on statements referencing internal paths (state/requirements.jsonl, logs/review.jsonl) and undefined modes (Lite/Full), with low-signal lines like the evidence_files sentence. Only the first sentence is well front-loaded; the rest is hard to parse in a single read.

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?

An output schema exists, so return values need not be described, which helps. Still, for a 4-parameter finishing/mutation tool with zero schema coverage and no annotations, the description omits what the terminal state actually emits, what artifact/task mean, and how the Lite checklist maps to submission, leaving real gaps despite the volume of text.

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. It explains message (intended user-visible delivery text) and checklist semantics well (template when omitted; submit passed/evidence items), but leaves task and artifact undescribed, and references an evidence_files field that does not exist in the schema, which misleads more than it clarifies.

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 opening sentence gives a specific verb and resource: verify report delivery and publish its receipt/terminal state when checks pass. That distinguishes it from research_status (read state) and research_start/save. However, terms like 'Lite', 'Full', and 'terminal state' are undefined and assume prior context, so the precise scope is not fully self-evident.

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?

It offers some when-to-use cues: omit checklist to get a template, submit passed/evidence items, append finding_disposition rows for evidenced no-change decisions, and evaluation tasks end via research_review. But these are tangled across internal file references rather than stated as a clean decision rule, and no explicit exclusion tells the agent when NOT to call this versus research_review/research_check_draft.

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