AgentTrust

Overview Schema Source Related Servers Discussions Admin

Evaluate Escrow Work

evaluate_escrow_work

Destructive

Submit proof of completed work to an escrow vault. Validates on-chain evidence and auto-releases payment on approval, or returns specific reasons for rejection.

Instructions

Submit proof of completed work against an existing escrow vault.

On PASS, payment releases automatically — no EscrowFinish needed. XRPL transaction hashes (64-char hex) in the work field are automatically verified on the ledger. Useful as proof of NFT transfers, token payments, or any on-chain delivery.

Response shape on PASS:

{
  "verdict":          "PASS",
  "status":           "approved",
  "score":            0-100,
  "summary":          "explanation",
  "criteria_met":     ["all criteria that passed"],
  "criteria_failed":  [],
  "auto_finish_queued": True,
  "model_used":       "gemini-2.5-flash"
}

Response shape on FAIL:

{
  "verdict":           "FAIL",
  "status":            "rejected",
  "score":             0-100,
  "summary":           "explanation",
  "criteria_met":      ["criteria that passed"],
  "criteria_failed":   ["specific unmet criterion — exact reason"],
  "attempts_remaining": 2,
  "model_used":        "gemini-2.5-flash"
}

On FAIL: read criteria_failed for specific, actionable feedback. Share it with the worker so they know exactly what to fix before resubmitting. Resubmit by calling evaluate_escrow_work again with the same escrow_id.

Input Schema

TableJSON Schema

Name	Required	Description	Default
`work`	Yes	Work submission or proof of completion. XRPL tx hashes (64-char hex) are auto-verified on the ledger.
`escrow_id`	Yes	The receipt code provided by the buyer when creating the vault.
`task_category`	No	Evaluation rubric. One of: default, creative, code, data, data_analysis, bug_bounty, legal, supply_chain.	default
`evidence_links`	No	Up to 3 URLs that are fetched and snapshotted at submission time as supporting evidence.
`require_consensus`	No	Require two AI models to agree before returning PASS. Recommended for high-stakes jobs.

Output Schema

TableJSON Schema

Name	Required	Description	Default
No arguments

Schema Changelog

Changes observed during successful MCP inspections.

First observedv0.1.0Sep 7, 2026

TDQS

A4.3/5.0

Behavior5/5

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

Annotations indicate destructive behavior (destructiveHint=true) and non-read-only, but the description goes far beyond that. It discloses that PASS triggers automatic payment without EscrowFinish, that XRPL transaction hashes are auto-verified, and that FAIL includes attempts_remaining and actionable feedback. It also explains the effect of require_consensus (not in description but in schema) implicitly through the PASS/FAIL flow. This is rich behavioral context.

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 long but justified by the tool's complexity: it includes two distinct response shapes, a note about automatic verification, and usage feedback instructions. Every sentence serves a purpose, and the key point (submit proof) is front-loaded. It is not bloated, though it could be slightly tightened by moving response shapes to the output schema definition.

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

Completeness4/5

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

Given the tool's complexity (5 parameters, detailed output behavior, annotations), the description is largely complete. It provides full response structures for PASS and FAIL, explains ledger verification, and gives clear next-step guidance for failures. Minor gaps exist—such as not explicitly stating how evidence_links are used or what task_category affects—but these are covered in the schema descriptions. The output schema is embedded in the description, so agents have enough context.

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

Parameters3/5

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

All five parameters are fully described in the input schema (100% coverage), so the schema already carries the documentation burden. The description adds no additional parameter-level meaning beyond what the schema states—for example, it does not elaborate on task_category or evidence_links beyond their schema descriptions. This meets the baseline for high schema coverage.

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 clear, specific action: 'Submit proof of completed work against an existing escrow vault.' The verb ('submit proof') and resource ('escrow vault') are unambiguous, and it immediately distinguishes this tool from escrow creation or confirmation tools. The note about automatic payment release further clarifies its role.

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

Usage Guidelines4/5

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

The description gives strong contextual guidance: it explains when to use this tool (after work is done, against an existing escrow) and how to act on results (share FAIL feedback, resubmit with same escrow_id). However, it does not explicitly contrast with sibling tools like audit_task or confirm_escrow_transaction, leaving some alternative-selection inference to the agent.

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