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

Annotations are sparse (all three hints false), so the description carries the burden, and it delivers: review routing ('routes the immutable claim to automatic, operator, or professional review'), the critical no-auto-completion behavior, idempotency retry guidance, canonicality advice ('trust the returned actual_tool_output ... instead of adding a state-recovery call'), and prerequisites. However, there is an internal tension: the claim is called 'immutable' yet the confirmation-boundary boilerplate classifies the action as a 'reversible save,' and 'plus every prerequisite stated above' is a vacuous reference to prerequisites that are never actually listed.

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 first two sentences front-load the essential semantics efficiently, but roughly half the description is templated boilerplate: the 'Canonicality,' 'Idempotency,' and 'Confirmation boundary' paragraphs read like generic footer text, with the confirmation boundary listing unrelated action categories ('read, reversible save, ... link preparation, plan refresh'). The prerequisite line is nearly tautological ('plus every prerequisite stated above'). The useful content is padded by filler that could be trimmed without losing information.

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 13-parameter, 11-required tool with nested objects and no output schema, the description covers intent, the most important behavioral caveat (evidence submission does not mark work complete), review routing, retry semantics, and trust guidance for the returned output. It is incomplete on the return payload (what the 'returned review' structurally contains), on constructing evidence/source objects, and on the role of _corply_context, which is only described in the schema. Strong for its complexity class, but not exhaustive.

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?

With schema description coverage at only 8%, the description must compensate, and it partially does: it explains the composite key semantics ('pins rule/version/subject/occurrence'), the role of the idempotency key ('durable idempotency key'), and attestation ('explicit attestation'). But the complex nested parameters — source, evidence items, professional, provenance, claimedOutcome — receive no semantic gloss, and the relationship between claimedOutcome and the subsequent review flow is only hinted at. The description adds meaning for roughly a third of the parameters while the rest rely on self-documenting schema structure.

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 states a specific verb, resource, and scope: 'Record evidence that the company completed one exact materialized work occurrence outside Corply.' It further pins down exactly what the command locks ('rule/version/subject/occurrence') and explicitly separates itself from completion-marking tools with 'This tool never marks the work completed merely because evidence was submitted.' An agent can distinguish it from siblings like transition_operating_work_item or record_operating_fact without opening their schemas.

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 gives explicit when-not guidance: 'use the normal fact and work-transition tools only after the returned review and remaining evidence/outcome gaps are resolved,' which maps directly to sibling tool categories. It also states the prerequisite condition (authenticated active organization access) and clarifies that this tool records evidence for work completed outside Corply, implying it is not for originating work. The routing-to-review behavior further tells the agent what to expect.

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