← WebMCP Directory

Tool tree

prism.alx21.chatgpt.site

Developer ToolsTools verified
Website
Evaluate a WebMCP implementation against the job it is actually designed to do.
Explore capabilities
    get_evaluation_contextAnswer

    Read the current draft profile, actual custom intent and tool names, approval rule, target label, and source selection shown in Prism.

    View tool JSON
    {
      "name": "get_evaluation_context",
      "kind": "answer",
      "impl": "imperative",
      "description": "Read the current draft profile, actual custom intent and tool names, approval rule, target label, and source selection shown in Prism.",
      "inputSchema": {
        "type": "object",
        "properties": {},
        "additionalProperties": false
      }
    }
    choose_evaluation_profileAction

    Change the visible draft contract. Preserve imported JSON and the previous report; that report is marked stale until evaluated again.

    View tool JSON
    {
      "name": "choose_evaluation_profile",
      "kind": "act",
      "impl": "imperative",
      "description": "Change the visible draft contract. Preserve imported JSON and the previous report; that report is marked stale until evaluated again.",
      "inputSchema": {
        "type": "object",
        "properties": {
          "profile": {
            "type": "string",
            "enum": [
              "commerce",
              "operations",
              "editor",
              "custom"
            ],
            "description": "Choose a built-in contract, or custom to use the custom fields currently entered."
          }
        },
        "required": [
          "profile"
        ],
        "additionalProperties": false
      }
    }
    run_sample_evaluationAction

    Evaluate a built-in synthetic example against the requested contract and visibly mark it as an example. Custom uses the current custom fields; this never tests a live target.

    View tool JSON
    {
      "name": "run_sample_evaluation",
      "kind": "act",
      "impl": "imperative",
      "description": "Evaluate a built-in synthetic example against the requested contract and visibly mark it as an example. Custom uses the current custom fields; this never tests a live target.",
      "inputSchema": {
        "type": "object",
        "properties": {
          "profile": {
            "type": "string",
            "enum": [
              "commerce",
              "operations",
              "editor",
              "custom"
            ],
            "description": "Choose a built-in contract, or custom to use the custom fields currently entered."
          }
        },
        "required": [
          "profile"
        ],
        "additionalProperties": false
      }
    }
    evaluate_current_snapshotAction

    Evaluate the JSON currently entered in Runner snapshot against the visible contract and update the report. Requires Runner snapshot mode. Evidence is supplied, not independently verified.

    View tool JSON
    {
      "name": "evaluate_current_snapshot",
      "kind": "act",
      "impl": "imperative",
      "description": "Evaluate the JSON currently entered in Runner snapshot against the visible contract and update the report. Requires Runner snapshot mode. Evidence is supplied, not independently verified.",
      "inputSchema": {
        "type": "object",
        "properties": {},
        "additionalProperties": false
      }
    }
    get_latest_evaluationAnswer

    Read the visible report including its exact contract, target, example or snapshot provenance, findings, and stale flag. A stale report does not describe the current draft.

    View tool JSON
    {
      "name": "get_latest_evaluation",
      "kind": "answer",
      "impl": "imperative",
      "description": "Read the visible report including its exact contract, target, example or snapshot provenance, findings, and stale flag. A stale report does not describe the current draft.",
      "inputSchema": {
        "type": "object",
        "properties": {},
        "additionalProperties": false
      }
    }