# Captchas Openclaw

> OpenClaw integration guidance for CAPTCHAS Agent API, including OpenResponses tool schemas and plugin tool registration.

- **Type:** Skill
- **Install:** `agentstack add skill-kernel8901-ai-agent-skills-classification-captchas-openclaw`
- **Verified:** Yes — security-reviewed for prompt injection and unsafe behavior
- **Seller:** [Kernel8901](https://agentstack.voostack.com/s/kernel8901)
- **Installs:** 0
- **Category:** [Agent Skills](https://agentstack.voostack.com/c/agent-skills)
- **Latest version:** 0.1.0
- **License:** MIT
- **Upstream author:** [Kernel8901](https://github.com/Kernel8901)
- **Source:** https://github.com/Kernel8901/ai-agent-skills-classification/tree/main/APIs与集成/captchas-openclaw

## Install

```sh
agentstack add skill-kernel8901-ai-agent-skills-classification-captchas-openclaw
```

Requires the [AgentStack CLI](https://agentstack.voostack.com/docs/cli). Works with Claude Code, Cursor, and any MCP-compatible agent.

## About

# CAPTCHAS + OpenClaw

Use this skill when integrating CAPTCHAS with OpenClaw via OpenResponses tools or OpenClaw plugin tools.

## Configuration

Set environment variables:

- `CAPTCHAS_ENDPOINT` = `https://agent.captchas.co`
- `CAPTCHAS_API_KEY` = ``

Headers:

- `x-api-key`: required (use `CAPTCHAS_API_KEY`).
- `x-domain`: optional; validated if provided.

Notes:

- `site_key` is optional; if omitted, it resolves from the API key or account default.
- Avoid sending PII in `signals`.

## OpenResponses Tool Schemas (OpenClaw Gateway)

Use the OpenClaw `tools` array shape when calling the Gateway `/v1/responses` endpoint.

```json
{
  "tools": [
    {
      "type": "function",
      "function": {
        "name": "captchas_agent_verify",
        "description": "Run CAPTCHAS Agent Verify and return a decision (allow|deny|challenge).",
        "parameters": {
          "type": "object",
          "properties": {
            "site_key": {"type": "string"},
            "action": {"type": "string"},
            "signals": {"type": "object", "additionalProperties": true},
            "capabilities": {
              "oneOf": [
                {"type": "object", "additionalProperties": true},
                {"type": "array", "items": {"type": "string"}}
              ]
            },
            "verification_mode": {"type": "string", "enum": ["backend_linked", "agent_only"]},
            "challenge_source": {"type": "string", "enum": ["bank", "ai_generated"]},
            "input_type": {"type": "string", "enum": ["choice", "image", "behavioral"]},
            "media_url": {"type": "string"},
            "media_type": {"type": "string"}
          },
          "required": [],
          "additionalProperties": false
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "captchas_agent_challenge_complete",
        "description": "Complete a challenge and mint a verification token when passed.",
        "parameters": {
          "type": "object",
          "properties": {
            "challenge_id": {"type": "string"},
            "site_key": {"type": "string"},
            "answer": {"type": "string"}
          },
          "required": ["challenge_id", "answer"],
          "additionalProperties": false
        }
      }
    },
    {
      "type": "function",
      "function": {
        "name": "captchas_agent_token_verify",
        "description": "Verify an opaque CAPTCHAS token before completing a sensitive action.",
        "parameters": {
          "type": "object",
          "properties": {
            "token": {"type": "string"},
            "site_key": {"type": "string"},
            "domain": {"type": "string"}
          },
          "required": ["token"],
          "additionalProperties": false
        }
      }
    }
  ]
}
```

## OpenClaw Plugin Tool Registration

Register tools using `api.registerTool(...)` and the same JSON Schema parameters as above.

Example:

```js
api.registerTool({
  name: "captchas_agent_verify",
  description: "Run CAPTCHAS Agent Verify and return a decision (allow|deny|challenge).",
  parameters: {
    type: "object",
    properties: {
      site_key: { type: "string" },
      action: { type: "string" },
      signals: { type: "object", additionalProperties: true }
    },
    required: [],
    additionalProperties: false
  },
  async execute(_id, params) {
    return { content: [{ type: "text", text: JSON.stringify(params) }] };
  }
});
```

## References

- Use `/v1/agent/verify`, `/v1/agent/challenge/:id/complete`, and `/v1/agent/token-verify` as the canonical API calls.
- See `captchas-human-verification/SKILL.md` for workflow guidance.

## Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

- **Author:** [Kernel8901](https://github.com/Kernel8901)
- **Source:** [Kernel8901/ai-agent-skills-classification](https://github.com/Kernel8901/ai-agent-skills-classification)
- **License:** MIT

Install and usage instructions live in the source repository linked above.

## Pricing

- **Free** — Free

## Security capabilities

Automated source analysis of v0.1.0 — what this tool can access:

- **Network access:** no
- **Filesystem access:** no
- **Shell / process execution:** no
- **Environment & secrets:** no
- **Dynamic code execution:** no

*"Yes" means the capability is present in the source — more access means more to trust, not that it is unsafe.*


## Versions

- **0.1.0** — security scan: passed — Imported from the upstream source.

## Links

- Listing page: https://agentstack.voostack.com/l/skill-kernel8901-ai-agent-skills-classification-captchas-openclaw
- Seller: https://agentstack.voostack.com/s/kernel8901
- Browse the marketplace: https://agentstack.voostack.com/browse

---
Listed on AgentStack — the marketplace for AI agent skills and MCP servers. Every listing is security-reviewed. Creators keep 70%.