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

Annotations already declare openWorldHint=true, idempotentHint=false and destructiveHint=false, so the description's job is to add what the annotations can't. It does: exact price ($0.05 USDC on Solana), the fact that a retry is a fresh paid call requiring the manifest to be re-sent, and that submitted content is processed in memory and never stored. The remaining gap is credible failure/payment behaviour — what happens if payment succeeds but the fetch fails, or whether the score is deterministic.

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?

Front-loaded with the price, which is the single decision-relevant fact, then the deliverable, then privacy, retry semantics and limits. Every sentence carries information. It is one dense run-on block rather than separated clauses, which slightly hurts scannability, but nothing is wasteful.

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?

There is no output schema, so the description must carry the return contract — and it does, itemising verdict/score, per-tool per-check results with evidence, findings categories (spec-default traps, both-claimed traps, schema inconsistencies) and a fix list. Combined with the explicit non-guarantees, an agent knows both what it gets and what it must not claim from the result.

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 coverage is 100%: both `url` (with GET, 512 KB / 10 s caps) and `manifest` (accepted shapes, in-memory only) are fully documented, including their mutual exclusivity. The description only restates the url-or-manifest choice in the retry sentence, adding no format or constraint detail beyond the schema. Baseline 3 is correct.

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 states a specific verb and resource — a paid audit of one submitted MCP manifest for one price — and enumerates exactly what the deliverable contains (verdict line and score, per-tool/per-check results with evidence, findings list, fix list). It also sharply bounds the resource: 'It does not execute the server, observe its runtime behaviour, or certify that the server is safe.' An agent can tell this apart from manifest_precheck/manifest_criteria/manifest_drift without opening any schema.

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 implied rather than stated: the phrase 'the paid retry re-sends the manifest (or its URL)' hints there is a cheaper first pass, but no sibling is named and no explicit 'use X first' or 'use Y instead for drift' routing is given. The exclusions describe scope boundaries, not alternatives. Adequate, but the agent must infer when this is the right tool versus manifest_precheck.

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