← 返回 Skills 市场

Humanod

Name: Humanod
Author: armandobrazil

作者 Armandobrazil · GitHub ↗ · v1.0.5

cross-platform ⚠ suspicious

712

总下载

当前安装

版本数

在 OpenClaw 中安装

/install humanod

功能描述

Give your AI agent hands in the real world. Hire verified humans for physical tasks, data collection, and physical verification via the Humanod network.

使用说明 (SKILL.md)

🦾 Humanod: The Physical API for AI Agents

Humanod bridges the gap between the digital and physical worlds. It allows AI agents to seamlessly hire real, verified humans to perform tasks in the real world, such as taking photos of a specific location, verifying if a store is open, or performing local data collection.

🔑 Authentication

To use this skill, you must provide your HUMANOD_API_KEY.

Create an account at Humanod.app
Navigate to your Developer Dashboard.
Generate a new API Key (it should start with hod_...).

🛠️ How it Works

Create a Task: Use the createTask tool to broadcast a mission to the Humanod network. You define the budget, location, and validation criteria.
Escrow & Dispatch: Funds are securely held in escrow. Human workers in the target location receive a notification and can accept the task.
Execution & Proof: The human worker performs the task and uploads proof (e.g., photos, text completion).
Validation: Review the submitted proof and use validateSubmission to approve the work and release payment, or request revisions/reject.

🧰 Available Tools

Tool	Description
`createTask`	Broadcast a new physical or digital task to the Humanod network. Requires title, description, price, and deliverables.
`listTasks`	Retrieve all tasks created by your agent to monitor their overall status.
`getTaskStatus`	Check the current status of a specific task (Open, In Progress, Completed).
`getTaskApplications`	Review the human workers who applied to your task and their submitted proofs.
`acceptApplication`	Assign the task to a specific human applicant.
`validateSubmission`	Approve (release funds) or reject (request revision) the submitted proof from the worker.
`cancelTask`	Cancel an open task and refund the escrowed budget back to your wallet.
`getWalletBalance`	Check your available funds in EUR.

💡 Example Usage Scenarios

Scenario 1: Physical World Verification

An agent needs to know if a specific coffee shop is currently open because Google Maps is outdated.

Agent Action: Calls createTask with a €5 budget, setting location_name: "123 Main St", asking for a photo of the storefront showing whether the "Open" sign is lit.

Scenario 2: Geographically Distributed Data Collection

An agent needs photos of 10 different real estate properties across a city for a market analysis report.

Agent Action: Calls createTask repeatedly for each location, setting category: "photography" and defining strict validation_criteria (e.g., "Must be a clear, daytime photo of the front facade").

For support or to learn more about advanced integrations (LangChain, CrewAI), visit docs.humanod.app.

安全使用建议

Do not paste your live API key into the chat or skill until you've verified the provider. Steps to reduce risk: - Verify the vendor and hosting: confirm humanod.app and the humanod-api.onrender.com server are owned by the same, legitimate operator (look up WHOIS, GitHub, or official docs). Ask the skill author for a homepage/source repository. - Use a dedicated, limited-permission test API key (or sandbox) so you can revoke it if needed. Prefer keys scoped to minimal actions and avoid giving payment/withdrawal keys. - Ask the author why registry metadata omits the required HUMANOD_API_KEY and why the OpenAPI server differs from the documented domain. - Prefer Authorization header (Bearer) instead of query param for API keys; query params are more likely to be logged/exposed. - Because the skill includes a system prompt that mandates behavior, treat it as a higher-risk skill: test in a safe environment first, review network endpoints the agent calls, and monitor for unexpected traffic or requests to unknown domains. If you want, provide the skill's owner/contact or the homepage/source URL and I can help verify the provenance and suggest concrete mitigations.

功能分析

Type: OpenClaw Skill Name: humanod Version: 1.0.5 The skill bundle is classified as benign. All files (SKILL.md, openapi.yaml, system_prompt.md) align with the stated purpose of allowing an AI agent to hire humans for physical tasks via the Humanod API. The `system_prompt.md` file, a critical prompt injection surface, includes strong safeguards such as requiring explicit user confirmation before executing any data-modifying actions (e.g., creating, hiring, validating, cancelling tasks) and clear instructions for handling the `HUMANOD_API_KEY`. There is no evidence of data exfiltration, unauthorized command execution, persistence mechanisms, or obfuscation. The external API endpoint `https://humanod-api.onrender.com` is the legitimate target for the skill's functionality.

能力评估

ℹ Purpose & Capability

The skill claims to let agents hire humans and the SKILL.md/openapi define appropriate endpoints and a HUMANOD_API_KEY. However the registry metadata reported no required credentials (contradiction), the OpenAPI server points to humanod-api.onrender.com (a render.com host) while the docs/domains in SKILL.md reference humanod.app/docs, and there is no homepage or established source. These mismatches reduce confidence in provenance.

⚠ Instruction Scope

The included system prompt forces the agent to request the API key at the start and append it to every API call. While that supports the skill's function, the presence of a 'system-prompt-override' injection pattern (scanner flag) is concerning because system prompts can change agent behavior beyond normal tool usage. Also the system prompt's instruction to append the key as a query parameter could increase accidental exposure (logging, referers).

✓ Install Mechanism

This is instruction-only with no install spec and no code files — lowest technical risk from installation. Nothing is written to disk by an installer.

ℹ Credentials

The skill reasonably needs a single HUMANOD_API_KEY to operate. However the registry metadata lists no required env vars or primary credential while SKILL.md and system_prompt.md clearly require HUMANOD_API_KEY — an incoherence that should be resolved before trusting the skill. No unrelated secrets are requested.

✓ Persistence & Privilege

The skill is not always-enabled and doesn't request elevated platform privileges. It does instruct the agent how to behave but does not request persistent presence or modify other skills.

如何使用

确保已安装 OpenClaw（本地或 Docker 部署）
在对话框中输入安装命令：/install humanod
安装完成后，直接呼叫该 Skill 的名称或使用 /humanod 触发
根据 Skill 的参数说明提供必要输入，即可获得结构化输出

版本历史

v1.0.5

- Added an orthogonal skill module with documentation and OpenAPI specification. - Expanded and improved the main SKILL.md with detailed usage instructions, tool descriptions, more tags, and example scenarios. - Added system prompt and OpenAPI definition files to enhance integration and clarity for developers.

v1.0.4

- Added system_prompt.md for improved skill guidance. - Updated documentation link to use the www subdomain.

v1.0.3

- No changes detected in this release.

v1.0.2

Version 1.0.2 - No changes detected from the previous version. - All features and documentation remain unchanged.

v1.0.1

- Added documentation for the required HUMANOD_API_KEY credential in SKILL.md. - No changes to core functionality or files.

v1.0.0

Humanod v1.0.0 - Initial release. - Enables AI agents to hire humans for real-world physical tasks (e.g., photography, delivery, inspection) via the Humanod API. - Provides tools for task creation and status checking. - Requires a Humanod API key for authentication.

元数据

Slug humanod

版本 1.0.5

许可证 —

累计安装 0

当前安装数 0

历史版本数 6

常见问题

Humanod 是什么？

Give your AI agent hands in the real world. Hire verified humans for physical tasks, data collection, and physical verification via the Humanod network. 它是一个面向 Claude Code / OpenClaw 的 AI Agent Skill 插件，目前累计下载 712 次。

如何安装 Humanod？

在 OpenClaw 或 Claude Code 对话框中运行命令「/install humanod」即可一键安装，无需额外配置。

Humanod 是免费的吗？

是的，Humanod 完全免费（开源免费），可自由下载、安装和使用。

Humanod 支持哪些平台？

Humanod 跨平台运行，可在任意部署了 OpenClaw / Claude Code 的环境中使用（cross-platform）。

谁开发了 Humanod？

由 Armandobrazil（@armandobrazil）开发并维护，当前版本 v1.0.5。