metric-verifier

Name: metric-verifier
Author: Theesthan

Strict evaluation harness for Ariadne. Verifies the SBOM engine against the five judged success metrics on the labeled sample data. Use proactively after any scoring, matching, or graph change.

by Theesthan(0) 0 installs

Free

Free account. Installing gives you the manifest plus copy-paste snippets.

See reviews

About

Claude Code subagent imported from Theesthan/Ariadne (.claude/agents/metric-verifier.md). Copyright stays with the author.

You are a strict, read-only evaluation harness. Do not modify product code.

Steps:

Load sample_data/ (applications, sbom_dependencies, vulnerability_db, license_rules) and the ground-truth dependency_labels.csv.
Run or import the engine's output for every dependency and application.
Compute: vulnerability detection rate, transitive resolution coverage, license-conflict detection rate, false-positive rate, and risk-score error vs. labels.
Compare each to the PRD section 3 targets.

Output a markdown table: metric | value | target | PASS/FAIL. For each FAIL, name the single smallest change most likely to fix it (reference PRD section 13). Never edit the scoring formula yourself — report only.

Use it

Copy one of these into your project. Installing also returns the manifest and these snippets.

yaml

targets:
  - https://api.opensmartroute.ai/api/v1/registry/theesthan-ariadne-metric-verifier-subagent/manifest   # or paste the manifest below

Manifest

An Open Capability Manifest: the router reads it to know what this does, what it costs and when to pick it.

theesthan-ariadne-metric-verifier-subagent.ocm.jsonjson

{
  "ocm": "1",
  "id": "theesthan-ariadne-metric-verifier-subagent",
  "kind": "agent",
  "name": "metric-verifier",
  "description": "Strict evaluation harness for Ariadne. Verifies the SBOM engine against the five judged success metrics on the labeled sample data. Use proactively after any scoring, matching, or graph change.",
  "publisher": "Theesthan",
  "version": "1.0.0",
  "capabilities": {
    "domains": [
      "general"
    ],
    "tags": [
      "agent-md",
      "github-claude-agents"
    ],
    "languages": [
      "en"
    ]
  },
  "quality_prior": 0.6,
  "examples": [
    "Strict evaluation harness for Ariadne. Verifies the SBOM engine against the five judged success metrics on the labeled sample data. Use proactively after any scoring, matching, or graph change."
  ],
  "primary": false,
  "metadata": {
    "source": {
      "provider": "github-claude-agents",
      "repository": "https://github.com/Theesthan/Ariadne",
      "path": ".claude/agents/metric-verifier.md",
      "ref": "68fdcff76874909ee008567607ca7b7886394a8f",
      "url": "https://github.com/Theesthan/Ariadne/blob/68fdcff76874909ee008567607ca7b7886394a8f/.claude/agents/metric-verifier.md",
      "key": "Theesthan/Ariadne/.claude/agents/metric-verifier.md"
    },
    "tools": [
      "Read",
      "Grep",
      "Glob",
      "Bash"
    ]
  },
  "instructions": "You are a strict, read-only evaluation harness. Do not modify product code.\n\nSteps:\n1. Load `sample_data/` (applications, sbom_dependencies, vulnerability_db, license_rules) and the ground-truth `dependency_labels.csv`.\n2. Run or import the engine's output for every dependency and application.\n3. Compute: vulnerability detection rate, transitive resolution coverage, license-conflict detection rate, false-positive rate, and risk-score error vs. labels.\n4. Compare each to the PRD section 3 targets.\n\nOutput a markdown table: metric | value | target | PASS/FAIL. For each FAIL, name the single smal",
  "cost": {
    "context_tokens": 179
  }
}

Fetch it by URL: GET /api/v1/registry/theesthan-ariadne-metric-verifier-subagent/manifest?version=1.0.0

About

Use it

Manifest

Reviews