Pre-open repo agent-surface audit

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

With no annotations, the description carries the full burden. It usefully discloses that the verdict is precomputed and signed and that each call costs $0.01 unless covered by a plan, but says nothing about auth requirements, behavior on a missing key, staleness, or what the signed payload contains.

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?

A single front-loaded sentence with a compact argument list and no filler. The pricing aside is relevant cost context rather than padding.

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 lookup tool with no output schema and no annotations, the description does not explain the returned verdict shape, failure behavior, or how it relates to verify_receipt. An agent would have to discover these by trial.

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%, so the description should compensate, but it only restates the three parameter names. It does not clarify that tree_hash is required while harness and version are optional, nor whether the key must match exactly.

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 a specific resource, a precomputed signed verdict, and the composite key it is keyed on (tree_hash x harness x version). An agent can tell this is a cached lookup rather than a computation, which distinguishes it somewhat from verify_receipt, though no sibling is named.

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?

There is no statement of when to use this versus preopen_audit or verify_receipt, nor what to do when no precomputed verdict exists. The pricing note gives context but not routing guidance.

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

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

With no annotations at all, the description carries the full disclosure burden, and it discloses almost nothing: no side effects, no indication whether the tree is mutated or merely inspected, no auth/plan requirements beyond price, and no sense of what a verdict looks like. The pricing detail ($0.99 after free trial, or plan-included) is genuinely useful behavioral context, which keeps this above a 1.

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 compact sentences with no filler, and the operation name plus the nested-args hint are front-loaded. It is tight, though the terse "Args:" notation compresses so far that meaning is lost for three of the four parameters.

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?

This is a complex tool with a nested object parameter, 0% schema coverage, no annotations, and no output schema. Given that the description is the sole source of semantics, it should explain the audit criteria, the meaning of harness/version/policy, and roughly what comes back; instead it supplies only a tree shape and a price.

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 0%, so the description must compensate. It does expand `tree` into files with path and one of content/content_b64/link, which is real value the schema (object, no detail) does not provide. But `harness`, `version`, and `policy` — including the required `harness` and `version` — receive zero explanation of expected values or 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?

"Pre-open audit of a received tree" gives a verb (audit) and a resource (a received file tree), and it is distinguishable from siblings like sanitise_tree and verify_receipt by name. However, "pre-open" and "received tree" are unexplained jargon, and the description never says what the audit actually checks or returns, so the agent knows only the general category of operation.

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?

There is no statement of when to call this rather than sanitise_tree, verify_receipt, or lookup_verdict, nor any prerequisite, ordering, or exclusion. The only contextual sentence is a pricing clause, which informs cost decisions but not tool selection.

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

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

No annotations are provided, so the description carries the full behavioral burden. It usefully discloses that the operation is metered ($1.99, always metered) and implies a quarantined copy, but it omits whether the copy is persisted, what gets quarantined, failure behavior, and permission requirements.

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 very short and front-loads the operation and cost. However, the trailing 'Args: tree, harness, version, policy' list duplicates the schema without adding meaning and consumes space that could have clarified behavior.

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?

With four parameters, nested tree and policy objects, no annotations, no output schema, and no parameter descriptions, the definition is far too thin for an agent to invoke correctly. It does not explain what tree or policy should contain, nor what the sanitised quarantined copy actually returns.

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 the description merely repeats parameter names ('tree, harness, version, policy') without explaining what any of them mean. For a tool with nested objects like tree and policy, this leaves all four parameters semantically undefined.

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 says 'Audit + sanitised quarantined copy' but never states a clear verb+resource outcome, leaving it ambiguous whether the tool audits, creates a copy, or both. It does not distinguish itself from the sibling preopen_audit, which also sounds audit-related.

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?

There is no guidance on when to use this tool versus lookup_verdict, preopen_audit, or verify_receipt. The only usage-like detail is 'always metered', which is pricing context rather than a decision rule.

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

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

No annotations are provided, so the description carries the full burden and does not meet it. It discloses one behavioral fact (the call is free) but says nothing about what verification yields, what happens on an invalid signature (error vs. false result), or any auth/permission requirements.

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?

It is short and front-loads the core purpose, which is good. However, the trailing 'Args: signed_verdict' clause adds zero information and the brevity comes at the cost of under-specification rather than efficient density.

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 tool with a nested-object input, no output schema, and no annotations, the description leaves too much unspecified: the receipt object's shape, the return value, and failure behavior are all unaddressed.

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 0% and the single parameter is a nested object with no documented properties. The description's 'Args: signed_verdict' merely repeats the parameter name and gives no insight into the structure or expected contents of the signed receipt object.

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 a specific verb ('Verify') and a specific resource ('Ed25519-signed receipt'), which is meaningfully distinct from siblings like lookup_verdict or sanitise_tree. It does not, however, explicitly contrast itself with lookup_verdict, which an agent could easily confuse with verification.

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?

There is no statement of when to use this versus lookup_verdict or preopen_audit, nor any prerequisite context. The only hint is '(free)', which implies a cost trade-off but never explains the alternative or the condition for choosing this tool.

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