Synthex Forge — x402 Paid Agent Services

Evaluate Task Delegation Safety

evaluate_task_delegation_safety

Read-onlyIdempotent

A well-formed delegation is not the same as a safe one. use it rather than accept_task_deliverable for task delegation safety, not task deliverable; an object with 41 fields including agent_id, at. $0.4 USDC.

Input Schema

Name	Required	Description
`commitments.agent_id`	No	[POST /v1/autonomy/commitment-check]
`delegation.delegation`	No	[POST /v1/autonomy/delegation-evaluate]
`delegation.delegate_id`	No	[POST /v1/autonomy/delegation-evaluate]
`commitments.commitments`	No	[POST /v1/autonomy/commitment-check]
`delegation.delegator_id`	No	[POST /v1/autonomy/delegation-evaluate]
`commitments.observations`	No	[POST /v1/autonomy/commitment-check]

Output Schema

Name	Required	Description
`at`	No	The `at` field of the folded answer.
`id`	No	The `id` field of the folded answer.
`ok`	No	Whether every inner call succeeded.
`due`	No	The `due` field of the folded answer.
`raw`	No	The `raw` field of the folded answer.
`kind`	No	The `kind` field of the folded answer.
`note`	No	The `note` field of the folded answer.
`text`	No	The `text` field of the folded answer.
`type`	No	The `type` field of the folded answer.
`state`	No	The `state` field of the folded answer.
`detail`	No	The `detail` field of the folded answer.
`reason`	No	The `reason` field of the folded answer.
`states`	No	The `states` field of the folded answer.
`status`	No	The `status` field of the folded answer.
`outcome`	No	The `outcome` field of the folded answer.
`results`	No	One result per inner call.
`summary`	No	One-line summary of the folded answer.
`agent_id`	No	The `agent_id` field of the folded answer.
`evidence`	No	The `evidence` field of the folded answer.
`dimension`	No	The `dimension` field of the folded answer.
`exception`	No	The `exception` field of the folded answer.
`dimensions`	No	The `dimensions` field of the folded answer.
`commitments`	No	The `commitments` field of the folded answer.
`delegate_id`	No	The `delegate_id` field of the folded answer.
`delegator_id`	No	The `delegator_id` field of the folded answer.
`match_method`	No	The `match_method` field of the folded answer.
`met_fraction`	No	The `met_fraction` field of the folded answer.
`observations`	No	The `observations` field of the folded answer.
`state_counts`	No	The `state_counts` field of the folded answer.
`commitment_id`	No	The `commitment_id` field of the folded answer.
`kept_fraction`	No	The `kept_fraction` field of the folded answer.
`observation_id`	No	The `observation_id` field of the folded answer.
`commitment_text`	No	The `commitment_text` field of the folded answer.
`dimension_order`	No	The `dimension_order` field of the folded answer.
`signals_matched`	No	The `signals_matched` field of the folded answer.
`commitment_count`	No	The `commitment_count` field of the folded answer.
`observation_text`	No	The `observation_text` field of the folded answer.
`observation_count`	No	The `observation_count` field of the folded answer.
`checkable_fraction`	No	The `checkable_fraction` field of the folded answer.
`observation_outcome`	No	The `observation_outcome` field of the folded answer.
`self_reported_status`	No	The `self_reported_status` field of the folded answer.

Schema Changelog

Changes observed during successful MCP inspections.

AddedOct 31, 2026

TDQS

B3.3/5.0

Behavior4/5

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

Annotations already declare readOnlyHint, idempotentHint, non-destructive behavior, and closed-world scope. The description adds useful context by disclosing the $0.4 USDC cost and rough output shape (an object with 41 fields including agent_id), which goes beyond structured metadata.

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

Conciseness2/5

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

The description is short but poorly structured, with sentence fragments and an awkward trailing phrase ("including agent_id, at. $0.4 USDC."). The first sentence is useful, but the rest does not read cleanly or front-load information well.

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

Completeness3/5

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

The tool has an output schema, full input-schema coverage, and safety annotations, so the description need not explain return structure in detail. However, it does not explain the evaluation criteria or the role of the nested delegation and commitments parameter groups, leaving meaningful context gaps.

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?

Schema description coverage is 100%, so the six parameters are structurally documented, but their descriptions are only endpoint tags such as [POST /v1/autonomy/delegation-evaluate]. The tool description adds no meaning for those parameters, so baseline 3 is appropriate.

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

Purpose3/5

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

The description signals the domain ("task delegation safety") and distinguishes it from accept_task_deliverable, but it never cleanly states what the evaluation produces or what makes a delegation unsafe. The fragmented sentence structure weakens the purpose statement.

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?

It explicitly says to use this rather than accept_task_deliverable for task delegation safety, not task deliverable. That gives an agent a clear alternative and selection condition, though it does not cover other possible alternatives.

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

All Synthex Forge — x402 Paid Agent Services 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.