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

The description goes well beyond the annotations (readOnlyHint=true, idempotentHint=true, destructiveHint=false). It discloses transient input processing (not stored/logged/retained), synthetic input requirement, browser delegation behavior for gpu nodes, and the FV-status snapshot semantics. It also clarifies the deterministic nature and the export artifact with execution_hash for provenance. No contradiction with annotations; description is highly transparent about behavioral traits.

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 description is dense and long, covering many aspects (compute modes, non-retention, synthetic data, artifact export, FV-status URL, hash, snapshot semantics). It front-loads the compute node role and compliance mandate but then presents several distinct facts with varying relevance. Some details seem tangential (exact URL for the artifact and FV-status hash) that could be omitted from the tool description for conciseness, though they might be valuable for chain provenance. The structure is logical but could be tightened by separating core behavior from provenance details.

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?

The description covers compute modes, side effects (non-retention, synthetic data requirement), the export artifact, and FV-status verification details. It lacks explicit output schema but denotes the export artifact with execution_hash. Since there's no output schema, the description does a reasonable job explaining what to expect. However, it references 'the tool's manifest' for policy_parameters fields, which is an external dependency that could be incomplete for the agent. Overall, quite complete given the complexity.

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%, so all four parameters are described in the schema already. The description adds context about compute='auto' server-side computation and browser delegation URL for 'browser' mode, matching the schema's enum description. For parent_hashes and parent_tool_ids, the description mentions 'chain parent_hashes in the export', but doesn't add much beyond the schema. Policy_parameters is described as 'the tool's decision function' inputs but defers to the manifest for field names, which is a partial gap. Overall, description contributes marginal value beyond schema, so baseline 3 is appropriate.

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?

States the tool checks SB 53 Frontier Scope and identifies it as an OpenChainGraph compute node. Describes compute mode behaviors, but doesn't explicitly distinguish from sibling tools like check_* functions (e.g., check_ssi_conformance, check_webbotauth_nonce_replay) beyond the SB 53 Frontier scope specificity. The 'compute node' framing with compliance_mandate and FV-status link provides context, though it's dense. Purpose is clear: a deterministic compute node for scope checking with configurable compute modes.

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?

Explains compute mode selection (auto/server/browser) and when browser delegation occurs, especially for gpu:true nodes. However, it does not explicitly state when to use this tool versus alternative check tools or sibling compute nodes. It implies usage through the FV-status and ChainGraph context, but lacks explicit exclusion guidance ('use X instead for Y'). The description is context-rich for compute behavior but weak on tool-vs-tool selection.

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