Skip to main content
Glama
hidai25

EvalView

by hidai25
OverviewSchemaRelated ServersScoreDiscussions
Python
Hybrid

run_skill_test

Execute skill test suites to validate AI agent performance through deterministic checks and rubric-based evaluation. Use for regression testing and quality assurance of coding agents.

Instructions

Run a skill test suite against a SKILL.md. Executes two evaluation phases: Phase 1 (deterministic) checks tool calls, file operations, commands run, output content, and token budgets. Phase 2 (rubric) uses LLM-as-judge to score output quality against a defined rubric. Call this after writing skill tests or after any change to the skill or agent. Use --no-rubric for fast Phase 1-only checks with no LLM cost.

Input Schema

TableJSON Schema
NameRequiredDescriptionDefault
test_fileYesPath to the skill test YAML file (e.g. 'tests/my-skill-tests.yaml')
agentNoAgent type to test against: 'claude-code', 'system-prompt', 'codex', 'langgraph', 'crewai', 'openai-assistants', 'custom'. Defaults to value in YAML.
no_rubricNoSkip Phase 2 rubric evaluation — run deterministic checks only (faster, no LLM cost). Default: false.
modelNoModel to use for evaluation (default: claude-sonnet-4-20250514)
verboseNoShow detailed output for all tests, not just failures. Default: false.

Implementation Reference

  • evalview/mcp_server.py:448-460 (handler)
    The logic handling the `run_skill_test` tool, which constructs and executes the `evalview skill test` CLI command.
    elif name == "run_skill_test":
    test_file = os.path.normpath(args.get("test_file", ""))
    if not test_file:
        return "Error: 'test_file' is required."
    cmd = ["evalview", "skill", "test", test_file, "--json"]
    if args.get("agent"):
        cmd += ["--agent", args["agent"]]
    if args.get("no_rubric") is True:
        cmd += ["--no-rubric"]
    if args.get("model"):
        cmd += ["--model", args["model"]]
    if args.get("verbose") is True:
        cmd += ["--verbose"]
  • evalview/mcp_server.py:204-229 (schema)
    The MCP tool registration and input schema definition for `run_skill_test`.
    {
    "name": "run_skill_test",
    "description": (
        "Run a skill test suite against a SKILL.md. "
        "Executes two evaluation phases: "
        "Phase 1 (deterministic) checks tool calls, file operations, commands run, output content, and token budgets. "
        "Phase 2 (rubric) uses LLM-as-judge to score output quality against a defined rubric. "
        "Call this after writing skill tests or after any change to the skill or agent. "
        "Use --no-rubric for fast Phase 1-only checks with no LLM cost."
    ),
    "inputSchema": {
        "type": "object",
        "required": ["test_file"],
        "properties": {
            "test_file": {
                "type": "string",
                "description": "Path to the skill test YAML file (e.g. 'tests/my-skill-tests.yaml')",
            },
            "agent": {
                "type": "string",
                "description": "Agent type to test against: 'claude-code', 'system-prompt', 'codex', 'langgraph', 'crewai', 'openai-assistants', 'custom'. Defaults to value in YAML.",
            },
            "no_rubric": {
                "type": "boolean",
                "description": "Skip Phase 2 rubric evaluation — run deterministic checks only (faster, no LLM cost). Default: false.",
            },

Tool Definition Quality

A4.4/5.0
Behavior4/5

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

With no annotations provided, the description carries full burden and does well by disclosing key behavioral traits: it explains the two-phase execution (deterministic checks and LLM-as-judge scoring), mentions performance implications ('faster, no LLM cost' for --no-rubric), and specifies when to call it. However, it doesn't cover error handling or output format details.

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?

The description is efficiently structured with three sentences: first states purpose and phases, second provides usage timing, third offers optimization guidance. Every sentence adds value with zero redundant information, making it appropriately sized and front-loaded.

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 5-parameter tool with no annotations and no output schema, the description provides good context about the tool's behavior, phases, and usage scenarios. It adequately compensates for the lack of structured fields, though it doesn't fully describe the output format or error conditions, which would be helpful given the complexity.

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 parameters thoroughly. The description adds minimal parameter semantics beyond the schema, only briefly mentioning '--no-rubric' in the context of Phase 1-only checks. This meets the baseline expectation when schema coverage is high.

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?

The description clearly states the specific action ('run a skill test suite') and resource ('against a SKILL.md'), detailing the two evaluation phases (deterministic checks and rubric scoring). It distinguishes from siblings by focusing on execution rather than creation (create_test, generate_skill_tests) or validation (validate_skill).

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?

Explicit guidance is provided on when to use the tool ('after writing skill tests or after any change to the skill or agent') and when to use alternatives ('--no-rubric for fast Phase 1-only checks with no LLM cost'). It clearly differentiates from other testing-related tools by specifying its execution role.

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

Install Server

Other Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/hidai25/eval-view'

If you have feedback or need assistance with the MCP directory API, please join our Discord server