Skip to main content
Glama
io.github.tammy357/ts-second-opinion

io.github.tammy357/ts-second-opinion

by tammy357

certify_trust

Calibrates conformal selection on half the RDB7 test reactions, then flags which remaining reactions can be trusted at a target error rate for geometry or handedness.

Instructions

Certified trust (conformal selection): calibrate on a random half of the RDB7 test reactions, then return which of the other half can be trusted so that, on average, at most target_error_rate of the trusted ones are wrong. count: 'geometry' (D-MAE > 0.2 A; what matters for barriers) or 'geometry_or_handedness'. The guarantee assumes new reactions are drawn like the calibration ones; recalibrate when the generator or data change.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "certify_trust"
  }
}

Input Schema

NameRequiredDescriptionDefault
seedNo
countNogeometry
target_error_rateNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.3.0

TDQS

A3.9/5.0
Behavior4/5

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

With no annotations, the description carries the full burden and does well: it explains the calibration/serve split, the statistical guarantee behind the returned trusted set, and the assumption the guarantee rests on. It omits cost/runtime, determinism relative to the seed, and the shape of the returned subset, leaving some gaps.

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?

Dense but front-loaded: the core operation, the count enum meaning, and the assumption/caveat each occupy one tight sentence. Nothing is decorative and the caveat is placed last where it belongs.

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?

For a three-parameter, no-annotation, no-output-schema tool, it covers the conceptual contract (what is returned and under what guarantee) adequately. The unexplained seed and the absence of any description of the returned structure are the remaining gaps, though the latter matters less for a subset-returning tool.

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 coverage is 0%, so the description must compensate, and it explains two of the three parameters meaningfully: count's values ('geometry' vs 'geometry_or_handedness') with the D-MAE > 0.2 Å criterion, and target_error_rate as the tolerated fraction of wrong trusted items. seed is left entirely unexplained, which keeps it short of a 5.

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?

Names a specific operation (conformal certification of trust) and describes concretely what it does: calibrate on a random half of the RDB7 test reactions and return which of the remaining half can be trusted. The procedure is distinctive enough that an agent can tell it apart from triage/evaluate tools, though no sibling is named explicitly.

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?

It gives real validity conditions — the guarantee assumes new reactions are drawn like the calibration ones, and it says to recalibrate when the generator or data change. However, it never states when to pick this over alternatives such as gate_check or evaluate_predictions, so the agent must infer the choice.

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