Skip to main content
Glama

DCL Trust Oracle — AI/LLM Output Audit (x402 MCP)

by com.fronesislabs

Batch Evaluation

dcl_evaluate_batch

PRE-ACTION Bulk Processing ($0.10). Evaluates a list of items in one call; each item is a dict shaped {"response": str, "policy"?: str}, where policy defaults to "default" if omitted and may be any built-in policy name (default, strict, anti_jailbreak, safety, content_quality). Each item gets its own independent COMMIT/NO_COMMIT verdict via the same logic as the matching single-item evaluate_* tool; results are returned in input order under results, plus a shared batch_id. Capped at 200 items per call — oversized batches are rejected. Use this instead of multiple single-item evaluate_* calls when checking several responses — optionally against different policies — in one priced call rather than paying per item separately.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
itemsYesList of items to evaluate, each shaped like {'response': str, 'policy'?: str}.
agent_idYesIdentifier of the agent that produced the responses.

Output Schema

TableJSON Schema
NameRequiredDescriptionDefault
countYesNumber of items evaluated in this batch.
resultsYesPer-item evaluation results, in input order.
agent_idYesIdentifier of the agent whose responses were evaluated.
batch_idYesUnique identifier for this batch run.

Schema Changelog

Changes observed during successful MCP inspections. Dates show when Glama detected each change.

  1. Changed1 schema field changed
    • removedInput schema / properties / payment_id
      Removed value: -{
-  "default": "",
-  "description": "Optional payment identifier returned by a previous call when payment is required",
-  "title": "Payment Id",
-  "type": "string"
-}
  2. First observed

TDQS

A4.9/5.0
Behavior5/5

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

Annotations only assert false hints, so the description carries the transparency burden and delivers: it discloses per-item independent COMMIT/NO_COMMIT verdicts, input-order result mapping, a shared batch_id, a 200-item cap with rejection behavior, and a $0.10 price. It also explains that each item follows the same logic as the corresponding single-item evaluate_* tool, which is useful behavioral context beyond 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.

Conciseness5/5

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

The description is front-loaded with the cost and bulk nature, then uses compact, purposeful clauses to cover item shape, policy options, verdict behavior, result order, output field names, cap, and usage compared to alternatives. Every sentence adds information; there is no filler or repetition of the title/schema verbatim.

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?

Given the tool's moderate complexity, the description supplies everything an agent needs to call it correctly: when to use it, item schema, policy defaults/options, per-item result semantics, output naming, and batch size limit. With an output schema also present, the absence of more detailed return documentation is not a gap.

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 description coverage is 100%, so the baseline is 3, but the description adds meaningful detail: the exact expected dict shape, the policy default of 'default', the allowed policy names, and the 200-item limit. It does not add anything for agent_id beyond the schema's own description, but the extra semantics on items clearly exceeds the bare schema.

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 states a specific action ('Evaluates a list of items in one call'), defines the item shape and verdict type, and explicitly distinguishes itself from the single-item evaluate_* tools by being the batch variant. It also names the output (`results` plus `batch_id`) and the 200-item cap, so an agent can identify its role among the many sibling evaluate tools.

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 an explicit usage rule: 'Use this instead of multiple single-item evaluate_* calls when checking several responses... in one priced call rather than paying per item separately.' The cap ('oversized batches are rejected') also states a hard limit, and the ability to mix policies clarifies when this batch tool is a better fit than policy-specific single-item siblings.

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

All DCL Trust Oracle — AI/LLM Output Audit (x402 MCP) toolsInspect
Try in Browser
RayrunBuild, deploy, and operate MCP servers without becoming an infrastructure team.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

TDQS

A3.7/5.0
Disambiguation3/5

There is meaningful overlap among the many dcl_evaluate_* variants (e.g., fast/strict/jailbreak/safety, secrets/wallet/pii, mev/signal/trade), so an agent could easily pick the wrong one. The descriptions are detailed and do help, but the boundaries between policies are subtle enough that the set is not clearly disambiguated.

Naming Consistency4/5

The dcl_evaluate_* family is highly consistent, and the remaining tools (dcl_commit, dcl_audit_decode, dcl_audit_decode_deep, dcl_pipeline_start) still follow the same snake_case dcl_ prefix convention. Minor deviations like using adjectives/adverbs in evaluate_fast/strict and noun-verb ordering in dcl_pipeline_start keep it from a perfect 5.

Tool Count3/5

18 tools is on the heavy side, especially since many are single-policy evaluators that could arguably be consolidated behind a policy parameter. That said, the tool count is justified by distinct pricing tiers and the breadth of audit categories, so it is borderline rather than excessive.

Completeness4/5

The audit lifecycle is well covered: pre-action evaluation, post-action scans, sanitization, final commit, and on-chain retrieval/verification. Minor gaps exist—no list/query tool for the append-only chain and no way to define custom policies—but the core workflows have no dead ends.

Resources