Skip to main content
Glama

veritas-8183-agent-notary-witness

audit_depin_gpu_sla_escrow

Idempotent

[Private] Is the rented GPU really what you paid for? Checks TFLOPS, VRAM, bandwidth and temperature against the contract. Example: minTflopsFp16=250, minVramGb=80.

Tool call template

Replace each placeholder with a value matching the input schema. Only required parameters are included; add optional parameters as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {
      "benchmarkReport": "<benchmarkReport>"
    },
    "name": "audit_depin_gpu_sla_escrow"
  }
}

Input Schema

NameRequiredDescriptionDefault
minVramGbNoMinimum required GPU VRAM in gigabytes.
minTflopsFp16NoMinimum required FP16 compute throughput in TFLOPS.
maxTempCelsiusNoMaximum safe GPU operating temperature in Celsius.
benchmarkReportYesHardware telemetry object with measuredTflopsFp16, availableVramGb, and gpuTempCelsius.
minMemoryBandwidthGbsNoMinimum memory bandwidth in GB/s.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed5 schema fields changed
    • addedInput schema / properties / benchmarkReport / description
      Added value: +"Hardware telemetry object with measuredTflopsFp16, availableVramGb, and gpuTempCelsius."
    • addedInput schema / properties / maxTempCelsius / description
      Added value: +"Maximum safe GPU operating temperature in Celsius."
    • addedInput schema / properties / minMemoryBandwidthGbs / description
      Added value: +"Minimum memory bandwidth in GB/s."
    • addedInput schema / properties / minTflopsFp16 / description
      Added value: +"Minimum required FP16 compute throughput in TFLOPS."
    • addedInput schema / properties / minVramGb / description
      Added value: +"Minimum required GPU VRAM in gigabytes."
  2. First observed

TDQS

B3.2/5.0
Behavior2/5

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

The annotations declare readOnlyHint=false, idempotentHint=true, destructiveHint=false, and openWorldHint=false, yet the description frames the tool purely as a check, saying nothing about what state it writes or what happens after a failed audit, such as an escrow verdict or a refund path. For a non-read-only escrow tool, that side effect is the most important thing to disclose. The '[Private]' marker and the list of measured dimensions add some context, but the write behavior implied by the annotations is left unexplained.

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 two short sentences plus an inline example, and it leads with the domain question before the mechanics, so it is front-loaded. The '[Private]' tag and the example earn their place. The question form is slightly indirect, but the description is not padded.

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?

There is no output schema, so the description should say what an agent gets back (pass/fail verdict, per-metric deltas, escrow consequence), but it does not. Given a nested required benchmarkReport and a non-read-only annotation profile, the definition leaves an agent unsure how to interpret or act on the result. It is adequate to invoke correctly, but thin for interpreting outcomes.

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 all five parameters, including the nested benchmarkReport fields, are already documented. The description adds example threshold magnitudes (minTflopsFp16=250, minVramGb=80), which is mildly useful for calibration, but it does not explain how thresholds are compared or what happens when a benchmarkReport is missing one of the telemetry fields. This meets the baseline expected when the schema does the heavy lifting.

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

Purpose4/5

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

The description states a concrete verb and scope: it verifies measured GPU telemetry (TFLOPS, VRAM, bandwidth, temperature) against contract minimums. That is specific enough to separate it from the other audit_*_escrow siblings, which cover API, inference, RAG, quant backtests, and so on. It stops just short of naming a sibling or explicitly contrasting the GPU case with those, so it is clear but not maximally differentiated.

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

Usage Guidelines3/5

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

The rhetorical opener ('Is the rented GPU really what you paid for?') and 'against the contract' imply the use case: validating a rented GPU against SLA thresholds. However, there is no explicit when-to-use/when-not guidance, no statement of prerequisites (for example, that a benchmarkReport must come from a trusted benchmark run), and no routing to alternatives with different SLAs.

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

Try in Browser

Glama MCP Gateway

Add one secure layer between your agents and this server.