SITEBORNE Utility Network

Verify agent output

siteborne_verify_agent_output

Idempotent

Evaluate a supplied agent output against explicit claims, deterministic requirements, a required JSON Schema, and optional evidence. Use when: the caller already has an output and needs a governed verification verdict. Do not use when: evidence must first be gathered from a company (use siteborne_build_company_evidence_graph), URL (use siteborne_retrieve_verified_web_context), or document (use siteborne_extract_document_evidence_json), or independent reproduction of the output is required (not available). Parameters: verification_mode must be standard; independent_reproduction is defined but unavailable and is rejected before any payment challenge, never downgraded to standard. required_schema must fit SITEBORNE JSON Schema Profile 1 or the request is rejected before payment; minimum_score sets the acceptance threshold; maximum_authorized_price only constrains payment and never relaxes verification policy. Behavior: closed-world: standard mode evaluates only the supplied claims, requirements, output and evidence with no outbound retrieval, and persists governed payment, audit, job and Workflow state; an unpaid request returns payment_required; a repeated identical request with the same payment is replay-protected and is not charged twice. The verification policy and verdict rules do not depend on price or maximum_authorized_price. This service is currently production-disabled and rejects execution. Economics: exact price of $0.017 USD per request in standard mode; the payment challenge requires that amount and the charge settles at exactly that amount. independent_reproduction mode has a governed price of $0.049 USD but is not available: it is rejected before any payment challenge and never substituted. Failure: an unsupported or oversized required_schema, schema-invalid input and unavailable modes are rejected before any payment challenge; a failed verification is a normal fail or conditional verdict, not an error. Returns: claim-level and deterministic-check results, an overall verification verdict, and PCC evidence—not newly gathered source content.

Input Schema

TableJSON Schema

Name	Required	Description
`minimum_score`	No	Minimum overall verification score from 0 through 1; raising it makes the final acceptance verdict stricter.
`required_schema`	Yes	JSON Schema the candidate output must satisfy; it constrains validation and cannot change SITEBORNE verifier policy.
`candidate_output`	Yes	Agent-produced JSON object to evaluate, bounded to 200 properties and depth 10; it is evidence under test, not trusted policy.
`supplied_evidence`	No	Up to twenty evidence objects supplied with the candidate output and bound to integrity hashes.
`verification_mode`	Yes	Verification mode. Only standard (deterministic checks of the supplied material, no outbound retrieval) is available. independent_reproduction has a governed price but is not available: it is rejected with verification_mode_unavailable before any payment challenge and is never downgraded to standard.
`verification_contract`	Yes	Complete set of claims and deterministic checks against which the supplied candidate output is evaluated.
`freshness_requirements`	No	Optional maximum-age limits for evidence and schemas, each expressed in seconds.
`allowed_evidence_sources`	No	Up to ten permitted evidence source URIs; evidence outside this allowlist cannot satisfy source restrictions.
`maximum_authorized_price`	No	Buyer’s decimal-safe maximum authorized price; constrains the payment requirement and never relaxes verification policy.

Output Schema

TableJSON Schema

Name	Required	Description	Default
`result`	Yes	Output schema for verify_agent_output.v1 service. Must be a valid PCC 1.0.0 document with required net.siteborne.agent-verification.v1 extension.

Schema Changelog

Changes observed during successful MCP inspections.

Changed1 schema field changedSep 20, 2026

changedInput schema / properties / verification_mode / description

Previous value: -"standard checks supplied material; independent_reproduction additionally attempts independently derived verification where implemented."New value: +"Verification mode. Only standard (deterministic checks of the supplied material, no outbound retrieval) is available. independent_reproduction has a governed price but is not available: it is rejected with verification_mode_unavailable before any payment challenge and is never downgraded to standard."

Changed26 schema fields changedSep 12, 2026

changedInput schema / properties / allowed_evidence_sources / description

Previous value: -"Allowed evidence source domains/URIs."New value: +"Up to ten permitted evidence source URIs; evidence outside this allowlist cannot satisfy source restrictions."

changedInput schema / properties / candidate_output / description

Previous value: -"The agent output to verify. Cannot modify verifier policy."New value: +"Agent-produced JSON object to evaluate, bounded to 200 properties and depth 10; it is evidence under test, not trusted policy."

addedInput schema / properties / freshness_requirements / description

Added value: +"Optional maximum-age limits for evidence and schemas, each expressed in seconds."

addedInput schema / properties / freshness_requirements / properties / evidence_max_age_seconds / description

Added value: +"Maximum evidence age in seconds from 0 through 2,592,000; older evidence fails freshness requirements."

addedInput schema / properties / freshness_requirements / properties / schema_max_age_seconds / description

Added value: +"Maximum schema age in seconds from 0 through 2,592,000; older schemas fail freshness requirements."

changedInput schema / properties / maximum_authorized_price / description

Previous value: -"Canonical decimal-safe money representation. No binary floating-point. Amount is a canonical decimal string."New value: +"Buyer’s decimal-safe maximum authorized price; constrains the payment requirement and never relaxes verification policy."

changedInput schema / properties / minimum_score / description

Previous value: -"Minimum verification score [0, 1]."New value: +"Minimum overall verification score from 0 through 1; raising it makes the final acceptance verdict stricter."

changedInput schema / properties / required_schema / description

Previous value: -"Required output schema (JSON Schema) that candidate must conform to."New value: +"JSON Schema the candidate output must satisfy; it constrains validation and cannot change SITEBORNE verifier policy."

changedInput schema / properties / supplied_evidence / description

Previous value: -"Evidence items supplied with the candidate output."New value: +"Up to twenty evidence objects supplied with the candidate output and bound to integrity hashes."

addedInput schema / properties / supplied_evidence / items / properties / content / description

Added value: +"JSON evidence content whose integrity must agree with content_hash; it is treated as untrusted input."

addedInput schema / properties / supplied_evidence / items / properties / content_hash / description

Added value: +"Lowercase SHA-256 integrity identifier in sha256:<64 hex> form for the supplied content."

addedInput schema / properties / supplied_evidence / items / properties / evidence_id / description

Added value: +"Caller-stable evidence identifier up to 64 characters, used to connect claims to this evidence item."

addedInput schema / properties / supplied_evidence / items / properties / retrieved_at / description

Added value: +"Optional UTC retrieval timestamp in the schema’s RFC 3339 form, used by freshness checks."

addedInput schema / properties / supplied_evidence / items / properties / source_uri / description

Added value: +"Optional public source URI, up to 2,048 characters, identifying where the evidence was obtained."

changedInput schema / properties / verification_contract / description

Previous value: -"Contract specifying what to verify."New value: +"Complete set of claims and deterministic checks against which the supplied candidate output is evaluated."

addedInput schema / properties / verification_contract / properties / claims / description

Added value: +"Up to fifty expected claims; each binds a stable claim identifier to a predicate, expected value, and optional tolerance/materiality."

addedInput schema / properties / verification_contract / properties / claims / items / properties / claim_id / description
```
Added value: +"Caller-stable claim identifier up to 64 characters, used to correlate the verdict for this claim."
```
addedInput schema / properties / verification_contract / properties / claims / items / properties / expected_value / description
```
Added value: +"Expected JSON value interpreted by the selected predicate; its shape should match the claim being tested."
```

addedInput schema / properties / verification_contract / properties / claims / items / properties / materiality / description

Added value: +"Claim importance—material, supporting, or contextual—which affects how the verdict weighs a failed claim."

addedInput schema / properties / verification_contract / properties / claims / items / properties / predicate / description

Added value: +"Comparison operation applied to the candidate value: equality, containment, pattern, ordering, existence, or nonexistence."

addedInput schema / properties / verification_contract / properties / claims / items / properties / tolerance / description
```
Added value: +"Optional nonnegative numeric tolerance for comparisons where bounded deviation is acceptable."
```

addedInput schema / properties / verification_contract / properties / deterministic_requirements / description

Added value: +"Up to thirty deterministic checks for schema, hashes, signatures, evidence resolution, PII, or secrets."

addedInput schema / properties / verification_contract / properties / deterministic_requirements / items / properties / check / description
```
Added value: +"Closed deterministic check to run: schema, hash, signature, evidence resolution, PII absence, or secret absence."
```
addedInput schema / properties / verification_contract / properties / deterministic_requirements / items / properties / parameters / description
```
Added value: +"Check-specific configuration interpreted only by the selected deterministic check; it cannot modify verifier policy."
```
addedInput schema / properties / verification_contract / properties / deterministic_requirements / items / properties / requirement_id / description
```
Added value: +"Caller-stable requirement identifier up to 64 characters, used to correlate the deterministic check result."
```

changedInput schema / properties / verification_mode / description

Previous value: -"Verification mode."New value: +"standard checks supplied material; independent_reproduction additionally attempts independently derived verification where implemented."

First observedSep 11, 2026

TDQS

A4.8/5.0

Behavior5/5

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

The description adds substantial behavioral context beyond annotations: closed-world no-outbound evaluation, persisted payment/audit/job/Workflow state, payment_required for unpaid requests, replay protection, production-disabled execution, exact pricing, and the fact that failed verification is a normal verdict rather than an error. These are exactly the non-obvious behaviors an agent needs to know.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness4/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is well structured with labeled sections (Use when, Parameters, Behavior, Economics, Failure, Returns) and front-loaded with the core purpose. It would earn a 5 if not for some repetition of the independent_reproduction unavailability and 'rejected before payment challenge' across multiple sections, though this repetition may be intentional given payment sensitivity.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a complex, paid, stateful verification service, the description is remarkably complete: it covers when to use it, when not to, payment behavior, replay protection, production-disablement, schema-profile constraints, failure semantics, and return contents. The output schema exists, so detailed return-value documentation is not the description's job.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema coverage is 100%, so the baseline is 3, but the description adds genuinely useful semantics beyond the schema: required_schema must fit SITEBORNE JSON Schema Profile 1, verification_mode must be standard while independent_reproduction is rejected, minimum_score sets the acceptance threshold, and maximum_authorized_price only constrains payment and never relaxes verification policy. Not every parameter is addressed in prose, but the schema already covers them.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description opens with a specific verb and resource: 'Evaluate a supplied agent output against explicit claims, deterministic requirements, a required JSON Schema, and optional evidence.' It clearly distinguishes this verification tool from sibling evidence-gathering tools by naming what it is not for, so an agent can select it without opening schemas.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It provides explicit use conditions ('Use when: the caller already has an output and needs a governed verification verdict') and explicit exclusions that name the correct sibling tools for evidence gathering from company, URL, or document. It also states that independent reproduction is not available, preventing a likely mis-selection.

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

All SITEBORNE Utility Network tools Inspect

Try in Browser

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources

Need Help?