Skip to main content
Glama

Run a saved model on the sealed test set and score it

evaluate

Runs a saved model on test features and scores its predictions against sealed labels, using one reveal; the model executes isolated with no network or label access.

Instructions

Run a saved model on the test features, then score its predictions on the sealed labels.

model_path is a joblib/pickle estimator, or a folder with predict.py defining predict(features) -> predictions. In sealed mode the model runs in an isolated container with no network and no access to labels. Spends one reveal when it succeeds.

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": {
      "dataset": "<dataset>",
      "model_path": "<model_path>"
    },
    "name": "evaluate"
  }
}

Input Schema

NameRequiredDescriptionDefault
seedNo
datasetYes
model_pathYes

Output Schema

NameRequiredDescriptionDefault
modeYes
stampNoProvenance: commit, dirty flag, data/scorer/lockfile hashes, seed, ledger entry
budgetNo
statusYes'scored' means metrics were revealed and one reveal was spent
datasetYes
messageYesPlain-English summary of what happened
metricsNo
findingsNo
n_scoredNo
remainingNoReveals left after this call

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observedv0.1.0

TDQS

B3.4/5.0
Behavior3/5

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 on the key traits: isolated container, no network, no label access, and a reveal is consumed on success. It omits what happens on failure (is a reveal still spent?), whether the run is idempotent, and any timeout or resource limits — meaningful gaps for a costly, sandboxed operation.

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 core action is front-loaded in the first sentence, followed by the model_path contract and the sealed-mode constraints. It is tight and free of filler, though the line-broken formatting slightly fragments the flow.

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?

An output schema exists, so return values need not be described. Still, for a three-parameter, high-cost tool with zero schema coverage, the definition leaves the dataset and seed parameters and the failure path unexplained, which is not quite enough for an agent to call it confidently.

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 0%, so the description must compensate. It explains model_path well (joblib/pickle estimator or a folder with predict.py defining predict(features) -> predictions), but 'dataset' and 'seed' receive no explanation at all, leaving two of three parameters undocumented in both schema and prose.

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 gives a specific two-stage verb chain: run a saved model on test features, then score predictions against sealed labels. That implicitly separates it from the sibling 'score' (which presumably scores supplied predictions rather than running a model), but the distinction is left for the agent to infer rather than stated.

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?

Usage context is implied by 'sealed test set' and the cost note 'Spends one reveal when it succeeds', which tells the agent this is a scarce, gated operation. However, there is no explicit when-to-use/when-not guidance and no named alternative for the case where the agent already has predictions.

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