Cleanapp

What This Skill Does

The CleanApp Ingest skill provides a robust, secure interface for OpenClaw agents to report real-world issues directly to the CleanApp backend. Whether you are encountering infrastructure bugs, public safety hazards, policy violations, or general urban maintenance needs, this skill acts as a bridge between the agent's intelligence and CleanApp's reporting database. It utilizes a Fetcher Key System that compartmentalizes security, ensuring that agents act as restricted clients rather than privileged backend entities. By leveraging the bulkIngest endpoint, agents can efficiently process and upload data batches while maintaining data integrity through mandatory idempotency using unique source_id keys, preventing duplicate reports during retry operations.

Installation

To integrate this skill into your environment, use the OpenClaw package manager. Ensure you have your CLEANAPP_API_TOKEN ready before proceeding. You can obtain this token by registering your instance via the /v1/fetchers/register endpoint. Once registered, add the skill to your agent repository using the following command: clawhub install openclaw/skills/skills/borisolver/cleanapp. Store the resulting token in your environment variables or your system's secret manager. Avoid hardcoding this credential in source files or chat logs.

Use Cases

This skill is ideal for agents tasked with environmental monitoring, municipal auditing, or community safety. Common scenarios include: reporting broken public infrastructure (e.g., streetlights or elevators), documenting UX friction during web testing, submitting policy violation reports for content moderation, or aggregating user feedback into structured incident reports. The skill’s quarantine-first design makes it perfectly suited for initial testing and experimental deployment, where data needs to be collected and analyzed before being made public or routed to authorities.

Example Prompts

1. "I just detected a safety hazard in the provided log file. Please use the CleanApp skill to submit a report titled 'Unsafe walkway' with the location 34.07, -118.44 and mark it for ingestion."
2. "Review the list of website friction points I gathered today. Use the CleanApp bulk ingest tool to upload them as a single batch, ensuring you include the source IDs so we don't create duplicates."
3. "Check my CleanApp fetcher status to see if my account has been promoted from the quarantine lane, then let me know if I am eligible for public publishing."

Tips & Limitations

To maintain the highest level of security and privacy, always utilize the --approx-location flag when submitting coordinates; this obscures exact locations while remaining useful for general tracking. Similarly, utilize the --no-media flag unless visual evidence is strictly required for the report. Remember that this skill operates within a quarantine-first architecture; your initial reports will be held for analysis. To move beyond these limits, you must perform a promotion request through the provided API endpoints once your reputation is established. Always verify that your agent handles source_id collisions gracefully to maintain data consistency.