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

With no annotations supplied, the description carries the full load and does disclose the critical behavioral facts: the operation is deterministic, it never pays a target, it does not adjudicate ('doesn't rule'), and it returns an object annotated with evidence strength and gaps. It also exposes deployment reality (MCP calls hit the AYNE playzone; production is a specific endpoint with a $0.002 x402 charge), which an agent needs to anticipate cost and environment. It loses a point for omitting permission/auth requirements, idempotency, and what happens when required evidence is missing.

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 core input-to-output mapping is front-loaded in a single dense sentence, followed by short scope disclaimers and a deployment note in parentheses. Nearly every clause earns its place, but 'By OMENZ.' is pure filler and the arrow/plus notation compresses so aggressively that it takes a second read to parse.

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?

Given no annotations, no output schema, and five fully undocumented parameters, the description does substantial work by describing the returned artifact ('signed dispute object with evidence strength and gaps') so return values need not be inferred. What remains missing is material: field-level formats, which parameters are required, and error behavior when evidence is incomplete. Adequate for an agent to recognize the tool's role, but not enough to construct a confident first call.

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% and every property is an empty object, so the description is the only source of parameter meaning. It productively maps four of the five parameters to concepts (order, failure, receipt_ids as 'payment receipts', snapshot_token as 'purchase snapshot'), which is real added value. But 'remedy' is never mentioned, no types or formats are given for any field (opaque token vs. ID list), and zero required parameters leaves the agent unsure what is mandatory.

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 arrow notation ('order + failure receipt + payment receipts + purchase snapshot -> a standard signed dispute object') states a concrete transformation from named inputs to a named artifact, which is far more specific than a restatement of the name. It also explicitly bounds the purpose by saying OMENZ 'packages evidence; it doesn't rule or move money.' It stops short of 5 because no sibling tool (e.g. finance_reconcile, finance_snapshot) is named as an alternative, so the agent must infer the boundary from context alone.

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?

Usage is only implied: the agent can infer this tool is for assembling refund/dispute evidence when an order, a failure receipt, payment receipts and a purchase snapshot are available. The 'doesn't rule or move money' clause functions as an implicit scope exclusion, signaling this is not the tool for actually issuing a refund. However, there is no explicit when-to-use statement, no prerequisites, and no named alternative for the actual payout step among the visible siblings.

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