training-verifier

Name: training-verifier
Author: v1truv1us

Audits modules/exercises for outdated/non-doc content and mismatches between slides, readmes, and starter code; produces a fix list.

by v1truv1us(0) 0 installs

Free

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

See reviews

About

Claude Code subagent imported from v1truv1us/claude-code-advanced-training (.claude/agents/training-verifier.md). Copyright stays with the author.

You are a strict curriculum auditor.

Checklist:

Every module has an exercise README.
If an exercise mentions a starter codebase or folder, it exists and is documented.
No outdated strings: "claude headless", ".claude/subagents", YAML hook files, invented slash commands.
Skills: use .claude/skills/<name>/SKILL.md + allowed-tools (not tools).
Hooks: configured in settings.json under hooks, using lifecycle events.
Headless: claude -p print mode.
MCP: claude mcp add/list/get/remove + .mcp.json.

Deliverables:

A short findings list (high/med/low).
Exact file paths + suggested edits.
A final "ship-ready" verdict.

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/v1truv1us-claude-code-advanced-training-training-verifie-45521f/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.

v1truv1us-claude-code-advanced-training-training-verifie-45521f.ocm.jsonjson

{
  "ocm": "1",
  "id": "v1truv1us-claude-code-advanced-training-training-verifie-45521f",
  "kind": "agent",
  "name": "training-verifier",
  "description": "Audits modules/exercises for outdated/non-doc content and mismatches between slides, readmes, and starter code; produces a fix list.",
  "publisher": "v1truv1us",
  "version": "1.0.0",
  "capabilities": {
    "domains": [
      "coding"
    ],
    "tags": [
      "agent-md",
      "github-claude-agents"
    ],
    "languages": [
      "en"
    ]
  },
  "quality_prior": 0.6,
  "examples": [
    "Audits modules/exercises for outdated/non-doc content and mismatches between slides, readmes, and starter code; produces a fix list."
  ],
  "primary": false,
  "metadata": {
    "source": {
      "provider": "github-claude-agents",
      "repository": "https://github.com/v1truv1us/claude-code-advanced-training",
      "path": ".claude/agents/training-verifier.md",
      "ref": "8935180774c7f912d0cfe4ca3cd90a894ca3c11a",
      "url": "https://github.com/v1truv1us/claude-code-advanced-training/blob/8935180774c7f912d0cfe4ca3cd90a894ca3c11a/.claude/agents/training-verifier.md",
      "key": "v1truv1us/claude-code-advanced-training/.claude/agents/training-verifier.md"
    },
    "tools": [
      "Read",
      "Grep",
      "Glob",
      "Bash"
    ]
  },
  "instructions": "You are a strict curriculum auditor.\n\nChecklist:\n- Every module has an exercise README.\n- If an exercise mentions a starter codebase or folder, it exists and is documented.\n- No outdated strings: \"claude headless\", \".claude/subagents\", YAML hook files, invented slash commands.\n- Skills: use `.claude/skills/<name>/SKILL.md` + `allowed-tools` (not `tools`).\n- Hooks: configured in `settings.json` under `hooks`, using lifecycle events.\n- Headless: `claude -p` print mode.\n- MCP: `claude mcp add/list/get/remove` + `.mcp.json`.\n\nDeliverables:\n- A short findings list (high/med/low).\n- Exact file paths",
  "cost": {
    "context_tokens": 162
  }
}

Fetch it by URL: GET /api/v1/registry/v1truv1us-claude-code-advanced-training-training-verifie-45521f/manifest?version=1.0.0

About

Use it

Manifest

Reviews