Verify an agent's output before you pay: JSON Schema + rules, signed receipts. Free during launch.
Scored across 2 tools
get_verification_record (retrieve a reputation/trust record) and verify_schema (run structural/content validation on an output) target clearly different actions despite both sitting under the 'verification' umbrella. The descriptions make the boundary explicit, though an agent skimming 'verify' vs 'get_verification' could momentarily hesitate.
Both names use snake_case verb_noun form (get_verification_record, verify_schema), which is predictable and readable. The verbs differ (get vs verify) but appropriately reflect retrieval vs validation rather than signaling inconsistency.
Two tools is thin for a server framed as an 'Agent Output Verifier,' which implies a broader verification workflow. Both tools earn their place, but the surface feels minimal rather than well-scoped.
Core needs are covered: check a counterparty's reputation and validate output before paying or submitting. However, the surface is read/check-only with no way to record, dispute, revoke, or batch-verify, leaving obvious lifecycle gaps for a trust/verification service.
Agent reputation and trust score: use before relying on another AI agent. Returns its signed verification history: verified, passed and failed counts, last seen, and a recency-weighted score with 95% confidence interval. Check a counterparty before you pay, or see the record buyers will see. AI Agent Scores: $0.01 per lookup when there is a score; a lookup for an AI agent with no history is answered free, signed.
| Name | Required | Description | Default |
|---|---|---|---|
| agent_id | Yes | The agent_id whose verification history to score (as passed to /verify/schema): the AI agent that produced the output checked by those calls (for example, the seller). agent_ids starting with 'test-' are reserved for testing and never have a history. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations cover the safety profile (readOnlyHint=true, destructiveHint=false), but the description adds material context annotations cannot: a per-lookup price of $0.01, a free lookup when no history exists, and that results are signed. It does not cover auth requirements or rate limiting, which keeps it short of 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?
Three tight sentences that lead with purpose and immediately follow with usage guidance. The pricing clause is appended at the end rather than front-loaded, but nothing is redundant or padded.
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 no output schema, the description carries the burden of describing the return shape, and it does so precisely (counts, last seen, weighted score, confidence interval) while also covering cost and signing. An agent has everything needed to decide and call correctly.
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 100% and the agent_id property already documents the identity semantics, the /verify/schema lineage, and the reserved 'test-' prefix. The description adds no parameter information beyond the schema, so the baseline 3 applies.
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 and resource ('Agent reputation and trust score', 'Returns its signed verification history') and enumerates the exact payload: verified/passed/failed counts, last seen, recency-weighted score with 95% CI. It is easy to distinguish from verify_schema conceptually, though the description itself never names the sibling to contrast with.
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?
Gives clear operating contexts: 'use before relying on another AI agent', 'Check a counterparty before you pay, or see the record buyers will see.' The when-to-use is concrete and actionable, but there is no explicit when-not or a named alternative such as verify_schema.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Independent acceptance check of AI agent output before paying for another AI agent's work, or before submitting your own: validate an invoice, check its total, or run JSON Schema validation. Checks structure, formats, ranges, and cross-field rules such as 'line totals must equal the total.' Returns pass/fail, % of checks passed, fix hints, and an Ed25519-signed attestation and receipt you can show as evidence of why you paid or refused. $0.02 per check, small next to the payment it protects. Free during launch: 1000 checks a day (MCP and REST).
| Name | Required | Description | Default |
|---|---|---|---|
| rules | No | Optional cross-field rules: sum_equals, gte, lte, exists_in, unique (duplicate detection). Violations are reported as informational flags unless enforce_rules is true. | |
| bounds | No | Optional per-field min/max, keyed by dotted path with '[]' array wildcard (e.g. 'items[].price'). Numbers or ISO-8601 dates. Violations are reported as informational flags unless enforce_rules is true. | |
| detail | No | 'full' (default) returns everything; 'compact' returns only the summary, next actions, receipt and signature. Use compact inside verify-repair loops to save tokens. 'compact' fields: summary, next_actions, result, the receipt fields (verification_id, task_id, agent_id, output_hash, schema_hash, rules_hash, verified_at, verifier_version, enforce_rules, strict_content_check), verify and attestation - errors, errors_detail, flags, hints and the scores are left out entirely, not just emptied. Both shapes are fully signed. | full |
| task_id | Yes | Your own identifier for this verification. 1-200 printable characters, no control characters and no lone Unicode surrogate; a request outside that range is rejected (422), not truncated. | |
| agent_id | No | Optional label identifying the AI agent that produced submitted_output (for example, the seller), used for /reputation and the trust score: records and scores belong to the producer, not to whoever calls this endpoint. It is an unauthenticated label. agent_ids starting with 'test-' are reserved for testing: the verification runs normally but is never recorded, so it cannot affect any reputation or trust score. 1-200 printable characters (same rule as GET /score/{agent_id}); a longer or malformed value is rejected (422), not truncated. | |
| enforce_rules | No | Opt-in. When true, every configured bound and consistency rule must hold, or the result is 'fail' (like strict_content_check), with an 'enforce_rules: ...' entry in errors. That includes a rule or bound that cannot run (its path is missing, a value is not comparable, a value has nothing to be compared against): it fails rather than passing silently, so omitting a field, in the whole path or in any array element the rule applies to, cannot skip a binding rule. A rule or bound with if_present: true is skipped where its field is absent instead. When false (default) nothing changes: violations and unrunnable rules stay informational flags. Echoed back in the signed response. | |
| expected_schema | Yes | Required. The JSON Schema (draft-04 to 2020-12) that submitted_output must satisfy: your own requirements, as a JSON object. | |
| submitted_output | Yes | Required. The output being checked: any JSON value (object, array, string, number, boolean or null), exactly as the AI agent produced it. | |
| strict_content_check | No | Opt-in. When true, any string in submitted_output containing control characters (e.g. null bytes) or embedded HTML/script markup makes the result 'fail'. When false (default) this is not checked at all. |
Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?
Annotations mark this as non-read-only and non-destructive, and the description adds real behavioral context beyond them: the $0.02 per-check cost, the 1000-checks/day free launch tier, and the Ed25519-signed attestation and receipt produced as evidence. That is meaningful cost, quota and side-effect disclosure, though it never explicitly says a verification is recorded against the producing agent's reputation (that only appears in the schema).
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?
Purpose, output and cost are front-loaded in the first sentences, and the structure moves cleanly from what it does to what it returns to what it costs. The promotional aside ('small next to the payment it protects') is the one sentence that leans more on persuasion than information.
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 nine parameters, nested rule/bound objects and no output schema, the description does the necessary work of sketching the return shape: pass/fail, percentage of checks passed, fix hints, and a signed attestation plus receipt. Pricing and quota are covered too. Only the reputation-recording consequence of a check is left to the schema.
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 100% and the nested defs (Bound, ConsistencyRule) carry thorough field-level documentation, so the baseline of 3 applies. The description only gestures at capability categories (structure, formats, ranges, cross-field rules) and adds no naming or syntax detail beyond what the schema already supplies.
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 opens with a specific verb+resource: an independent acceptance check that validates output against a JSON Schema, catching structure, formats, ranges and cross-field rules. Concrete examples (invoice validation, line totals equalling the total) make the operation unambiguous. It never names the sibling get_verification_record, so it stops short of full sibling differentiation.
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?
It clearly states the triggering context: run it before paying another agent, or before submitting your own work, and mentions both MCP and REST access. The 'compact' detail mode is flagged for verify-repair loops in the schema. It gives no explicit when-not-to-use guidance or a pointer to the retrieval sibling for follow-up steps.
Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.
Recent tool additions, removals, and schema changes observed during successful MCP inspections.
verify_schema2 fields changedInput schema / properties / expected_schema / descriptionAdded value: +"Required. The JSON Schema (draft-04 to 2020-12) that submitted_output must satisfy: your own requirements, as a JSON object."Input schema / properties / submitted_output / descriptionAdded value: +"Required. The output being checked: any JSON value (object, array, string, number, boolean or null), exactly as the AI agent produced it."get_verification_record1 field changedInput schema / properties / agent_id / descriptionPrevious value: -"The agent_id whose verification history to score (as passed to /verify/schema): the agent that produced the output checked by those calls (for example, the seller). agent_ids starting with 'test-' are reserved for testing and never have a history."New value: +"The agent_id whose verification history to score (as passed to /verify/schema): the AI agent that produced the output checked by those calls (for example, the seller). agent_ids starting with 'test-' are reserved for testing and never have a history."verify_schema1 field changedInput schema / properties / agent_id / descriptionPrevious value: -"Optional label identifying the agent that produced submitted_output (for example, the seller), used for /reputation and the trust score: records and scores belong to the producer, not to whoever calls this endpoint. It is an unauthenticated label. agent_ids starting with 'test-' are reserved for testing: the verification runs normally but is never recorded, so it cannot affect any reputation or trust score. 1-200 printable characters (same rule as GET /score/{agent_id}); a longer or malformed value is rejected (422), not truncated."New value: +"Optional label identifying the AI agent that produced submitted_output (for example, the seller), used for /reputation and the trust score: records and scores belong to the producer, not to whoever calls this endpoint. It is an unauthenticated label. agent_ids starting with 'test-' are reserved for testing: the verification runs normally but is never recorded, so it cannot affect any reputation or trust score. 1-200 printable characters (same rule as GET /score/{agent_id}); a longer or malformed value is rejected (422), not truncated."verify_schema1 field changedInput schema / $defs / ConsistencyRule / properties / tolerance / descriptionAdded value: +"Absolute tolerance for sum_equals and the near-match check gte/lte use. The default (1e-9) suits exact arithmetic but is usually too tight for money that passed through any floating-point math (e.g. percentage discounts, tax, unit-price * quantity): 0.1 + 0.2 != 0.3 in IEEE-754 floats by about 5.5e-17, and real pricing data regularly carries larger rounding drift than that. For a field in major currency units (dollars, euros), 0.01 (one cent) is a reasonable tolerance; for minor units (cents) already stored as integers, the default is fine. Too loose a tolerance can mask a genuine mismatch, so prefer the smallest value that tolerates your data's own rounding, not a large one 'to be safe.'"get_verification_recordverify_schemaIndependent schema and rule check of AI agent output before you pay. Signed attestation and receipt.
Paid agent work and paid verification, settled in USDC. Verified outcomes get signed receipts.
Issue signed receipts for AI agent actions; verify any receipt offline - free, no account.
Paid deterministic data-quality and execution-verification tools for AI agents.
Add one secure layer between your agents and this server.
