Back to plugin

Security audit

claw.cleaning

Security checks for vulnerabilities and agentic risk

Overview

The plugin's code, manifests, and SKILL.md consistently describe a booking skill that calls a public REST API at https://claw.cleaning; it requests no credentials or installs and its behavior matches its description.

This skill appears internally consistent, but installing it means the agent will send customer information (full address, name, email) to https://claw.cleaning. Before enabling: verify you trust the claw.cleaning service and its privacy practices; confirm you are comfortable that no upfront payments are taken (the flow relies on paying the cleaner at the appointment); keep the safety rule to always show a booking preview and require explicit confirmation before POSTing; if you prefer, do not allow autonomous invocation (or restrict the plugin) to prevent the agent from making bookings without explicit user approval. If you are concerned about the endpoint, consider auditing the remote service or only installing if the serverUrl points to a trusted host.

Static analysis

No suspicious patterns detected.