Skip to main content
Glama

ZeroWidth Caliper

by ai.zerowidth
OverviewDiscussionsAdmin

Run a Caliper eval

caliper_evals_run

Queues a new run of an eval — inference over the dataset, then LLM-judge scoring. THE VERIFY STEP of the improvement loop: after an approved workbench_flows_edit_text, run the eval again and report the score delta vs the previous run. Runs take a while — but you're brought back into THIS conversation automatically with the scores the moment it finishes, so tell the user it's queued and that you'll follow up here; never poll or ask them to check back. Costs workspace LLM budget, so it sits behind the approval gate: may return needs_confirmation.

Input Schema

NameRequiredDescriptionDefault
labelNoShort label for the run, e.g. 'after refund-policy fix'.
evalIdYesEval to run.
workspaceNoWorkspace slug. Personal tokens with no default workspace MUST pass this; tokens with a default can override per call. Ignored for workspace API keys.
approvalIdNoApproval id from a prior needs_confirmation envelope.
scenarioIdNoRun as this scenario instead of the eval's own (one of the flow's, from workbench_flows_scenarios_list); null runs with none. Omit to use the eval's. Running the same eval as two scenarios compares how the flow does for each person.

Schema Changelog

Changes observed during successful MCP inspections.

  1. Changed1 schema field changed
    • addedInput schema / properties / scenarioId
      Added value: +{
+  "anyOf": [
+    {
+      "type": "string"
+    },
+    {
+      "type": "null"
+    }
+  ],
+  "description": "Run as this scenario instead of the eval's own (one of the flow's, from workbench_flows_scenarios_list); null runs with none. Omit to use the eval's. Running the same eval as two scenarios compares how the flow does for each person."
+}
  2. First observed

TDQS

A4.6/5.0
Behavior5/5

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

Goes well beyond the annotations: discloses async/long-running behavior with automatic callback into the same conversation, cost against workspace LLM budget, and the approval gate that may return `needs_confirmation`. This is exactly the kind of behavioral context annotations cannot convey, and it is consistent with readOnlyHint=false / destructiveHint=false.

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?

Front-loaded with the core action and pipeline, then workflow role, then async/cost/approval caveats — every sentence is load-bearing. It is dense and relies on emphasis formatting, but there is no redundancy.

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?

For a mutating, asynchronous, cost-incurring tool with no output schema, the description covers what the agent cannot see elsewhere: return behavior (scores and delta, or a needs_confirmation envelope), the approval path, and follow-up expectations. Nothing needed to call it correctly is missing.

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 schema already documents all five parameters (label, evalId, workspace, approvalId, scenarioId) with examples. The description adds only indirect framing — the approvalId is implied by 'approval gate' — but no syntax or format detail beyond the schema. Baseline 3 applies.

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?

States a specific verb and resource ('Queues a new run of an eval') and even describes the internal pipeline — inference over the dataset, then LLM-judge scoring. This clearly distinguishes it from siblings like caliper_evals_runs_get, caliper_evals_runs_cancel, and caliper_evals_runs_list.

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?

Explicitly positions the tool as 'THE VERIFY STEP of the improvement loop' and names the triggering precursor (after an approved workbench_flows_edit_text), plus the goal (report score delta vs previous run). It also gives concrete do/don't guidance for the async case: tell the user it's queued and follow up, never poll. Nothing is left to inference.

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

All ZeroWidth Caliper toolsInspect
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.

Resources