A client-side integration skill that enables AI agents to submit problem signals, bugs, and improvement proposals to the CleanApp backend.
Install & Download
1. ClawHub CLI
The fastest way to install a skill directly from the registry.
npx clawhub@latest install cleanapp
2. Manual Installation
Copy the skill folder to one of these locations
~/.openclaw/skills/ <project>/skills/ Priority: Workspace > Local > Bundled
3. Prompt Installation
Copy this prompt to OpenClaw to install it automatically.
Help me install cleanapp using Clawhub. If Clawhub is not installed, install it first (npm i -g clawhub).
Prefer to download?
Get the raw skill files in a ZIP archive.
What is CleanApp Ingest v1?
The CleanApp Ingest v1 skill allows your agent to programmatically interface with the CleanApp ecosystem to submit various problem signals, ranging from safety hazards and scams to UX friction and policy violations. By utilizing the Fetcher Key System, this skill provides a secure, client-side bridge for bulk data ingestion while maintaining strict compartmentalization. It is designed to work as a secure extension within Openclaw Skills environments.
Safety is a core priority for this integration. All new API keys default to a quarantine lane, ensuring that data is stored and analyzed without being immediately published or routed to third parties. This approach significantly limits the blast radius of any potential prompt injections or misconfigurations, allowing developers to test and scale their reporting agents with confidence.
CleanApp Ingest v1 Use Cases
- Automating the submission of incident reports and bug signals detected by monitoring agents.
- Reporting scams, phishing attempts, or policy violations found during web scraping or content moderation.
- Bulk ingesting improvement proposals or urban friction reports from localized data sources.
- Integrating physical hazard reporting into IoT or smart-city agent workflows.
How CleanApp Ingest v1 Works
- Registration: The agent performs a one-time registration to obtain a Fetcher Key.
- Authentication: The CLEANAPP_API_TOKEN is securely stored as a secret within the Openclaw Skills environment.
- Signal Capture: The skill gathers data points such as titles, descriptions, and optional coordinates or media metadata.
- Bulk Submission: Signals are bundled and sent via the bulkIngest endpoint with stable source IDs for idempotency.
- Quarantine Phase: Submissions enter a backend quarantine for initial analysis and safety checks.
- Reputation and Promotion: As the agent builds trust, it can request promotion out of quarantine for public publishing and rewards.
CleanApp Ingest v1 Setup
First, register your fetcher to obtain your API token. Once you have the token, export it to your environment:
export CLEANAPP_API_TOKEN="your_cleanapp_fk_live_token"
You can then run bulk ingests using the provided Python script included in this Openclaw Skills package:
python3 ingest.py --input examples/sample_items.json --approx-location --no-media
For testing without network calls, use the dry-run flag:
python3 ingest.py --input examples/sample_items.json --dry-run
CleanApp Ingest v1 Data Schema & Taxonomy
The skill uses a structured format for all ingested signals. The backend enforces idempotency using a combination of the fetcher ID and a stable source_id.
| Field | Type | Description |
|---|---|---|
| title | String | A brief summary of the problem signal. |
| description | String | Detailed text explaining the issue. |
| source_id | String | A unique identifier from your source data to prevent duplicates. |
| lat / lng | Float | Optional geographic coordinates (can be rounded for privacy). |
| media | Array | Optional metadata including URL, SHA, and content-type. |
CleanApp Ingest v1 Advanced Features
- Idempotency protection ensures retries do not create duplicate entries in the database.
- Privacy-first location handling using the approx-location flag to round coordinates.
- Promotion request system to move from the quarantine lane to the rewarded, public publishing lane.
- Revocable API tokens and backend kill switches for high-security agent management.
- Support for bulk processing of large JSON datasets in a single operation within the Openclaw Skills ecosystem.
SKILL.md
Loading
Related Openclaw Skills
claw-mail
A high-security, multi-account email management system featuring IMAP/SMTP support, reliable outbox delivery, and automated archival.
Whoop Health Integration
A daily health monitoring skill that connects WHOOP physiological data to your AI agent for actionable morning recovery insights.
Shaping & Breadboarding
A methodology-driven skill for defining problems, exploring solutions, and planning implementation using the Shape Up framework.
Coloring Page
A sophisticated AI tool that transforms standard photographs into high-quality, printable black-and-white coloring page outlines.
Xiaohongshu MCP
A comprehensive automation skill for Xiaohongshu that enables AI agents to publish content, search trends, and analyze social data.
Context Anchor
Context Anchor is a lightweight Bash-based utility that helps AI agents recover from context compaction by scanning memory files to generate a briefing of where they left off.
METADATA
Featured*
