proofrelay.plan_replay_rejection_testReturn a deterministic replay rejection test plan and optionally classify observed first/replay responses. The tool does not redeem tokens or call paid endpoints.
| Name | Required | Description | Default |
|---|---|---|---|
| replay_test | Yes |
| Name | Required | Description | Default |
|---|---|---|---|
No arguments | |||
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations already declare readOnlyHint=true, idempotentHint=true, destructiveHint=false, so a safe read is established. The description adds genuine, agent-critical value by explicitly stating 'does not redeem tokens or call paid endpoints' — a payment-domain behavior not captured in annotations. It also discloses determinism. No contradiction. A slightly deeper disclosure on the optional classification branch would merit a 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?
Two sentences, roughly 25 words, front-loaded with the primary action and ending with the most important safety qualifier. Every clause earns its place with zero redundancy or filler.
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 the nested ReplayObservation schema, the payment-protocol enum, and the output schema, the description covers the two critical non-obvious aspects: the plan/classify dual modality and the absence of paid side effects. The presence of an output schema means return-value documentation is not the description's job. Only the thin framing of the tool's role in the broader payment-proof workflow keeps this from a 5.
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 0% schema description coverage, the description must compensate. It usefully clarifies that 'observed first/replay responses' maps to the nested ReplayObservation (first_status/second_status/error pairs) and clarifies the plan-vs-classify dual mode. However, it leaves the semantics of redeem_url, bundle_hash, entitlement_hash, and payment_protocol entirely unexplained — an agent cannot determine what these identifiers represent or when the enum values apply.
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 uses a specific verb+resource construct ('Return a deterministic replay rejection test plan') with a clear secondary behavior ('optionally classify observed first/replay responses') plus a key qualifier ('does not redeem tokens or call paid endpoints'). It cleanly distinguishes itself from its 26 build/verify/describe siblings, none of which cover replay rejection planning.
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 context is only implied through the word 'plan' and the qualifier 'deterministic.' The no-side-effects statement hints that this is the safe, no-cost planning alternative, but no explicit when-to-use, when-not-to-use, or sibling alternatives are named. The agent must infer when this non-redeeming planner takes precedence over actionable payment tools.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Add one secure layer between your agents and this server.
All 26 tools have distinct names that clearly indicate their purpose (e.g., adapt_agent_identity_evidence vs. build_audit_pack_manifest). No two tools have overlapping or ambiguous functions, even though some share similar verbs; the nouns and description contexts differentiate them effectively.
The tool names follow a uniform convention: all start with 'proofrelay.' and use lowercase verb_noun with underscores (e.g., proofrelay.build_bundle_draft, proofrelay.verify_signed_attestation). There is no mixing of camelCase, across the entire set.
The server exposes 26 tools, which exceeds the 'borderline' range of 16–25 and falls into the 'too many' category (>25). While the tools cover a broad domain (evidence verification, risk scanning, etc.), the large number might overwhelm agents and suggests that some tools could be merged or the scope narrowed.
The tool set covers a comprehensive range of operations—adapting, building, verifying, scanning, recommending, and summarizing—across multiple subdomains (identity, payment, audit, etc.). Minor gaps exist, such as missing tools for updating or deleting evidence, but overall the surface appears very complete for the described functionality.