LaunchKit · 2026
Back to Skills

cleanapp

This is a **skill package** that lets an agent submit **any problem signal** into CleanApp (bugs, incidents, scams.

0
786 downloads
by @borisolver

Setup & Installation

openclaw skills install @borisolver/cleanapp

Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:

npx clawhub install cleanapp

Version History

v0.1.1Feb 15, 2026 - Version note by skill developer:

CleanApp v0.1.1 - Updated release of the CleanApp skill package for submitting any type of bug report, issue, improvement proposal, hazard (in digital as well as physical worlds). - Supports report submission to CleanApp using the Fetcher Key System and quarantine-first backend. - Includes bulk ingest via `ingest.py`, sample item JSON, and shell helper scripts. - Documentation and API reference files provided for setup, usage, and best practices. - Secure by design: uses revocable API tokens and compartmentalized data handling. - Promotion process and backend safety/introspection endpoints detailed.

What This Skill Does

Submits reports to CleanApp via the Fetcher Key System. New submissions land in a quarantine lane, stored and analyzed but not publicly published or routed until promoted. Supports bulk ingest from JSON with optional coordinate approximation and idempotency via stable source IDs.

The quarantine-first model and revocable API key contain the blast radius of any agent misbehavior to quarantined-only submissions, unlike direct publishing integrations.

When to use it

  • Batch-uploading litter reports collected during a neighborhood cleanup
  • Ingesting 311-style civic complaints from a spreadsheet export
  • Testing a new data feed safely before requesting out-of-quarantine promotion
  • Submitting infrastructure damage reports with approximate GPS coordinates
  • Automating nightly report uploads from a field data collection app

Example Workflow

Here's how your AI assistant might use this skill in practice.

INPUT

User asks: ingest pothole reports from reports.json

AGENT
  1. 1Reads reports.json and validates required fields (title, description, source_id)
  2. 2Authenticates using CLEANAPP_API_TOKEN
  3. 3Calls POST /v1/reports:bulkIngest with --approx-location to reduce coordinate precision
  4. 4Checks response for idempotency conflicts and quota errors
  5. 5Returns ingest summary with accepted and quarantined item counts
OUTPUT

42 reports submitted, all quarantined pending review. 0 duplicates skipped.

Requirements

Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.

CLEANAPP_API_TOKEN — obtain by calling POST /v1/fetchers/register or via your CleanApp account