How do I install the Humanod skill?

Run npx skillmds@latest add javimosch/humanod in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Humanod safe to use?

This skill has not completed SkillMD's automated safety review yet. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Humanod?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Humanod free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Humanod?

javimosch (@javimosch) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Humanod by javimosch — Agent Skill

Name: Humanod
Author: javimosch

File contents

🦾 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.

---name: humanoddescription: Give your AI agent hands in the real world. Hire verified humans for physical tasks, data collection, and physical verification via the Humanod network.---# 🦾 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.## 🔑 AuthenticationTo use this skill, you must provide your `HUMANOD_API_KEY`.1. Create an account at [Humanod.app](https://www.humanod.app)2. Navigate to your Developer Dashboard.3. Generate a new API Key (it should start with `hod_...`).## 🛠️ How it Works1. **Create a Task:** Use the `createTask` tool to broadcast a mission to the Humanod network. You define the budget, location, and validation criteria.2. **Escrow & Dispatch:** Funds are securely held in escrow. Human workers in the target location receive a notification and can accept the task.3. **Execution & Proof:** The human worker performs the task and uploads proof (e.g., photos, text completion).4. **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 VerificationAn 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 CollectionAn 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](https://docs.humanod.app).*

Frequently asked questions

Humanod