middle_corridor_deal_risk

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

With no annotations provided, the description carries the full burden and exceeds it. It discloses exactly what the tool returns (triage recommendation, risk signal, decision-readiness score, supplied vs. minimum-required source categories, evidence gaps, high-risk-jurisdiction flag), and — critically — what it does NOT do (no live retrieval, no factual-truth verification, no legal/sanctions advice). It even routes the agent to get_schema for the full contract rather than hiding the nesting.

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?

Five dense, purposeful sentences, each earning its place: trigger points, input contract reference, output enumeration, and firm scope limitations. The disclaimers about no legal advice and human review are operationally necessary for a sanctions-adjacent tool and not padding. Loses one point for density — the single-sentence cascade of disclaimers could be lightly structured — but there is no appreciable waste.

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 high-complexity tool (nested request object, sanctions-adjacent domain, 1 param with 100% schema coverage, no output schema, no annotations), the description compensates thoroughly: it enumerates all six output components since no output schema exists, explains the input shape, and points to the full contract location. Loses a point because, without an output schema, the return-value prose list is the only contract and its structure/format remains unspecified — a minor gap given how much the description gets right.

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 100%, so the baseline is 3. The description adds value above baseline by enumerating the key fields inside deal_risk_request (route, cargo, counterparties, dated_sources, risk_question, decision_stage) and pointing to 'middle-corridor-deal-risk-request.schema.json' plus the get_schema call for the full nested contract. The schema's own description similarly guides the agent to get_schema('middle_corridor_deal_risk_request'). The only reason not to give a 5 is that the description's list is illustrative rather than exhaustive of the nested contract, so the agent must still fetch the schema for field-level semantics.

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 description opens with a specific verb + resource + scope: 'Screen a Kazakhstan / Middle Corridor (Trans-Caspian) trade deal for sanctions-adjacent and corridor risk.' It names exact trigger points (before signature, shipment, insurer handoff, committee review) and the corridor focus clearly distinguishes it from siblings like gulf_maritime_exposure and cis_secondary_sanctions_exposure without needing to compare 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 'when' is explicit and well-covered: 'before signature, shipment, insurer handoff, or committee review,' and the 'when-not' is also clear with 'Pre-compliance evidence triage only: no live retrieval, no factual-truth verification, no legal or sanctions advice; human review is required.' It misses a point only by not naming alternative sibling tools explicitly (e.g., when to prefer cis_secondary_sanctions_exposure instead), though the corridor-specific scope makes this largely self-evident.

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