← Back to Skills Marketplace
chloepark85

Post and manage errands for human workers

by Chloe Park · GitHub ↗ · v1.0.0
cross-platform ✓ Security Clean
672
Downloads
0
Stars
0
Active Installs
1
Versions
Install in OpenClaw
/install errand-ai
Description
Post errands using natural language, track status, review submissions, and automate USDC payments via the ErrandAI decentralized task marketplace.
README (SKILL.md)

ErrandAI Skill

Overview

Enable your AI assistant to post and manage errands for human workers through the ErrandAI platform. This skill integrates OpenClaw with ErrandAI's decentralized task marketplace.

Features

  • 🤖 Natural Language Commands - Post errands using conversational language
  • 📊 Status Tracking - Check errand status and submissions in real-time
  • Work Review - Approve or reject submissions with feedback
  • 💰 Automated Payments - USDC payments released automatically upon approval
  • 🌍 Global Reach - Post errands for any location worldwide

Installation

Prerequisites

  • OpenClaw v1.0.0 or higher
  • Node.js v14.0.0 or higher
  • ErrandAI API key (get from errand.be/dashboard)

Quick Install

openclaw skill install errand-ai

Manual Installation

  1. Download the skill files
  2. Copy to OpenClaw skills directory:
    cp -r errand-ai ~/.openclaw/skills/
    
  3. Set your API key:
    export ERRANDAI_API_KEY="your_api_key_here"
    
  4. Enable the skill:
    openclaw skill enable errand-ai
    

Configuration

Environment Variables

# Required
ERRANDAI_API_KEY=your_api_key_here

# Optional (defaults shown)
ERRANDAI_API_URL=https://api.errand.be

OpenClaw Config

# ~/.openclaw/config.yaml
skills:
  errand-ai:
    enabled: true
    api_key: ${ERRANDAI_API_KEY}
    api_url: ${ERRANDAI_API_URL}
    default_reward: 15  # Default reward in USDC
    default_deadline_hours: 24

Usage Examples

Post an Errand

You: Post an errand to check iPhone 15 stock at Apple Store Gangnam for $20
OpenClaw: ✅ Errand posted successfully!
Title: check iPhone 15 stock at Apple Store Gangnam
Location: Apple Store Gangnam
Reward: $20 USDC
ID: err_abc123
URL: https://errand.be/errand/err_abc123

Check Status

You: Check errand err_abc123
OpenClaw: 📋 Errand Status
Title: check iPhone 15 stock at Apple Store Gangnam
Status: in_progress
Reward: $20 USDC
Submissions: 2

Review Submissions

You: Approve submission sub_def456
OpenClaw: ✅ Submission approved! Payment has been released to the worker.

Supported Commands

Command Description Example
post errand Create a new errand "Post errand to take photo of menu at Starbucks for $15"
check errand Check errand status "Check errand err_123456"
list my errands List all your errands "Show my posted errands"
review submission Approve/reject work "Approve submission sub_789"

Natural Language Patterns

The skill understands various natural language patterns:

  • "Create an errand to..."
  • "I need someone to..."
  • "Post a task for..."
  • "Check the status of..."
  • "Approve/Reject submission..."

Categories Supported

  • 📸 Photography - Product photos, location verification
  • 🔍 Product Verification - Stock checks, availability
  • 💰 Price Research - Price comparisons, market research
  • 📝 Translation - Document and menu translation
  • 📊 Research - Surveys, interviews, data collection
  • 📦 Delivery - Package pickup, delivery confirmation
  • 🎯 General - Custom tasks

API Integration

Endpoints Used

  • POST /api/openclaw/errands - Create new errand
  • GET /api/openclaw/errands/{id} - Check errand status
  • POST /api/openclaw/submissions/{id}/review - Review submission
  • GET /api/openclaw/errands - List user's errands

Response Format

{
  "success": true,
  "errand": {
    "id": "err_abc123",
    "title": "Check iPhone stock",
    "status": "in_progress",
    "reward_amount": 20,
    "submissions_count": 2,
    "url": "https://errand.be/errand/err_abc123"
  }
}

Error Handling

The skill handles common errors gracefully:

  • Missing API key: Prompts to set ERRANDAI_API_KEY
  • Network errors: Retries with exponential backoff
  • Invalid commands: Provides helpful examples
  • API errors: Returns clear error messages

Security

  • API keys stored as environment variables
  • All API calls use HTTPS
  • Webhook signatures verified
  • No sensitive data logged

Troubleshooting

API Key Issues

# Verify API key is set
echo $ERRANDAI_API_KEY

# Test API connection
curl -H "X-API-Key: $ERRANDAI_API_KEY" https://api.errand.be/api/openclaw/health

Skill Not Loading

# Check skill status
openclaw skill status errand-ai

# Reload skills
openclaw skill reload

# Check logs
tail -f ~/.openclaw/logs/skills.log

Common Issues

Issue Solution
"API key not configured" Set ERRANDAI_API_KEY environment variable
"Failed to post errand" Check network and API status
"Errand not found" Verify errand ID format (err_xxxxx)
"Unauthorized" Check API key validity

Advanced Features

Bulk Operations

// Post multiple errands
"Post 3 errands for price checks at different stores"

Automated Workflows

// Schedule daily errands
"Every day at 9am, post errand to check coffee prices"

Custom Validation

// Set specific validation criteria
"Post errand with requirement: must include receipt photo"

Performance

  • Average response time: \x3C500ms
  • Concurrent errand limit: 10
  • Rate limit: 100 requests/minute
  • Webhook latency: \x3C100ms

Changelog

v1.0.0 (2024-02-14)

  • Initial release
  • Basic errand posting and management
  • Natural language processing
  • Submission review functionality
  • USDC payment integration

Support

License

MIT License - See LICENSE file for details

Contributing

We welcome contributions! Please see CONTRIBUTING.md for guidelines.

Credits

Created by the ErrandAI team for the OpenClaw ecosystem.

Usage Guidance
This skill is internally consistent: it calls ErrandAI endpoints and needs an ERRANDAI_API_KEY. Before installing: 1) Verify the API host (https://api.errand.be) and the origin of the skill (the registry metadata had inconsistencies; prefer installing from a trusted source such as the official repo). 2) Treat ERRANDAI_API_KEY like any API secret — set it in environment variables, not in code, and ensure it's limited in scope if possible. 3) Be aware the skill sends basic agent context (user id, channel, timestamp) to ErrandAI — if that contains sensitive identifiers you may want to avoid sharing them. 4) If you enable webhooks or scheduling/automation, audit the webhook endpoints you provide and verify signature verification and network exposure. 5) Test with small or no payments first to confirm behavior (automatic payment on approval is handled by the platform). 6) If you need stronger assurance, ask the publisher to correct the manifest/registry metadata mismatch and provide an authoritative source (homepage or GitHub repo) before widespread use.
Capability Analysis
Type: OpenClaw Skill Name: errand-ai Version: 1.0.0 The OpenClaw AgentSkills skill bundle for ErrandAI appears benign. All code (`errandai.skill.js`) and documentation (`SKILL.md`, `README.md`, `manifest.json`) align with the stated purpose of integrating with the ErrandAI platform to post and manage errands. The skill correctly handles API keys via environment variables or configuration, uses HTTPS for all external communication to `https://api.errand.be`, and declares necessary `network` and `api` permissions. There is no evidence of data exfiltration, malicious execution (e.g., `eval`, `exec`, `child_process`), persistence mechanisms, or prompt injection attempts designed to subvert the AI agent's behavior. Potential vulnerabilities like ReDoS in regex parsing or downstream API injection are not indicative of malicious intent within this skill.
Capability Assessment
Purpose & Capability
The code, SKILL.md, and manifest all implement posting, checking, and reviewing errands via ErrandAI endpoints and reference automated USDC payments handled by the platform — this is coherent with the skill's name/description. However, the registry summary at the top of the submission listed no required environment variables while the manifest and SKILL.md declare a required ERRANDAI_API_KEY; that metadata mismatch is inconsistent and should be corrected.
Instruction Scope
Runtime instructions and the code call ErrandAI API endpoints and instruct users to set ERRANDAI_API_KEY and optional ERRANDAI_API_URL. The skill includes examples for webhooks, scheduling, and bulk operations which can cause automated posting if you enable them. The skill includes metadata it sends to the API (openclawUser, channel, timestamp) — this is within scope for tracking but is user/agent context that will be transmitted to an external service and users should be aware of that data flow.
Install Mechanism
No remote downloads or installers; this is instruction + a small JS file that you copy into ~/.openclaw/skills/. No suspicious URLs or archives; risk from install mechanism is low.
Credentials
The manifest and SKILL.md require a single API key (ERRANDAI_API_KEY) and an optional API URL — proportionate to a remote API-integration skill. The only inconsistency is that the registry summary earlier claimed no required env vars; otherwise the requested environment variables and permissions (network/api) match the purpose. No unrelated credentials or host-level config paths are requested.
Persistence & Privilege
The skill is not marked always:true and will be user-invocable; it does not modify other skills or system-wide settings. It requires network permission to call ErrandAI APIs, which is expected for its function.
How to Use
  1. Make sure OpenClaw is installed (local or Docker)
  2. Run the install command in chat: /install errand-ai
  3. After installation, invoke the skill by name or use /errand-ai
  4. Provide required inputs per the skill's parameter spec and get structured output
Version History
v1.0.0
ErrandAI Skill v1.0.0 - Initial release integrating OpenClaw with the ErrandAI task marketplace - Supports posting errands with natural language commands - Real-time status tracking and submission review features - USDC payment automation upon approval of work - Global errand posting and management - Includes error handling, troubleshooting, and advanced automation options
Metadata
Slug errand-ai
Version 1.0.0
License
All-time Installs 0
Active Installs 0
Total Versions 1
Frequently Asked Questions

What is Post and manage errands for human workers?

Post errands using natural language, track status, review submissions, and automate USDC payments via the ErrandAI decentralized task marketplace. It is an AI Agent Skill for Claude Code / OpenClaw, with 672 downloads so far.

How do I install Post and manage errands for human workers?

Run "/install errand-ai" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.

Is Post and manage errands for human workers free?

Yes, Post and manage errands for human workers is completely free (open-source). You can download, install and use it at no cost.

Which platforms does Post and manage errands for human workers support?

Post and manage errands for human workers is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).

Who created Post and manage errands for human workers?

It is built and maintained by Chloe Park (@chloepark85); the current version is v1.0.0.

💬 Comments