WIA360: AI session probe and match

score_session

Submit the target AI session's raw reply text to a probe mission (from run_probe) for rule-based scoring. If the verdict is "invalid" or "omitted_pending", call run_probe again with the SAME session_ref to get a fresh variant of the same probe and retry once (pass the returned attempt_id back as retry_of) — after a second failure just move on, it is recorded as-is. IMPORTANT — lang must match the lang that was passed to the run_probe call that produced this mission; scoring uses a language-specific keyword dictionary and silently misfires if lang doesn't match how the mission was worded.

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": {
      "session_ref": "<session_ref>",
      "probe": "<probe>",
      "variant": "<variant>",
      "response_text": "<response_text>"
    },
    "name": "score_session"
  }
}

Input Schema

Name	Required	Description
`lang`	No	Must match the lang used in the run_probe call for this probe/variant. Defaults to ko.
`probe`	Yes
`variant`	Yes
`retry_of`	No	attempt_id from a prior invalid/omitted_pending result for this same probe, if this is a retry.
`session_ref`	Yes
`response_text`	Yes	The target AI session's raw reply text (for P4, the TURN2 reply only).

Schema Changelog

Changes observed during successful MCP inspections.

First observedOct 17, 2026

TDQS

A4.3/5.0

Behavior4/5

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

Annotations declare readOnlyHint=false, openWorldHint=true, idempotentHint=false, destructiveHint=false. The description adds important context: scoring uses a language-specific keyword dictionary and silently misfires if lang doesn't match. It also describes the retry workflow and 'invalid'/'omitted_pending' verdicts. That's valuable beyond annotations, though it doesn't mention authentication or rate limits.

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-loads the core action, then the retry flow, then the critical lang warning. Well-structured and information-dense. Slightly long but every sentence earns its place except perhaps the retry detail which could be condensed.

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?

Given no output schema and 50% schema coverage, the description provides the key operational rules: verdict handling, retry logic, and lang matching. It omits details about what the scoring returns (verdict values) and the meaning of probe/variant, but for a scoring submission tool the critical workflow is covered. A little more on return values would make it complete.

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 50%. The description explains lang (must match run_probe), retry_of (attempt_id from prior invalid/omitted_pending result), and response_text (raw reply, for P4 the TURN2 reply only). However, it does not explain probe, variant, or session_ref semantics, and those lack schema descriptions. Baseline 3 is appropriate given partial coverage and clear additions for some params.

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 (Submit) and resource (target AI session's raw reply text to a probe mission for rule-based scoring). Clearly distinguishes from run_probe (which produces the mission) and get_match. An agent can tell what this does without opening the schema.

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 says to retry once via run_probe with the same session_ref, pass attempt_id as retry_of, and after a second failure move on. Also includes a critical prerequisite: lang must match the lang passed to the run_probe call that produced the mission. This is unusually thorough usage guidance.

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

All WIA360: AI session probe and match tools Inspect

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

Need Help?