Qwen Agent

Interact with the Qwen agent. This skill wraps input in strict JSON enforcement prompts and handles retries.

Activation

Triggers: "ask qwen", "qwen help", "consult qwen"

Features

• Strict JSON Output: The agent is prompted to return ONLY valid JSON.
• Retry Logic: Automatically retries with stricter prompts if the first response is not valid JSON.
• Schema Support: Supports passing a JSON schema to constrain/shape the output (best-effort; enforced by instruction).

Usage

Simple Prompt

# Usage: .claude/skills/qwen/qwen-agent.sh "Your prompt here"
.claude/skills/qwen/qwen-agent.sh "Summarize the history of AI."

JSON Request (Advanced)

You can pass a JSON object with prompt, context, and schema.

{
  "prompt": "Generate a user profile",
  "context": "Target audience: internal admin tools",
  "schema": {
    "type": "object",
    "properties": {
      "username": {"type": "string"},
      "age": {"type": "integer"}
    },
    "required": ["username", "age"],
    "additionalProperties": false
  }
}

.claude/skills/qwen/qwen-agent.sh '{"prompt":"...","context":"...","schema":{...}}'

Response Format

Success (JSON only):

{
  "key": "value"
}

Error:

{
  "ok": false,
  "error": {
    "code": "error_code",
    "message": "Description"
  }
}

Environment Variables

• QWEN_CMD (default: qwen)
• QWEN_ARGS (default: empty)
• QWEN_TIMEOUT (default: 60)
• QWEN_APPROVAL_MODE (default: plan)