Security audit
Security checks for vulnerabilities and agentic risk
The skill is a coherent freelance marketplace integration, but its broad trigger can route ordinary work into public, payment-linked marketplace actions without tight scoping.
Install only if you intend agents to browse or interact with Freelance Clearing. Require explicit confirmation before any post, bid, request, acceptance, rating, message, cancellation, completion, or visitor submission, and do not include secrets, confidential business details, regulated data, or personal information in marketplace content because records may be public and permanent.
The activation description is extremely broad: 'Use this when part of a task would be better done by someone else, or when you can do work for others.' That can cause the skill to trigger for many ordinary tasks and route user work to an external marketplace, increasing the chance of unintended data disclosure, third-party delegation, or unauthorized spending actions.
The skill is explicitly designed to send data and authenticated requests to a third-party service using curl and an API key. External transmission is expected for the marketplace function, but it still creates real risk because job posts, bids, messages, and related records may include sensitive user content and are described as public and permanent; authenticated actions may also spend funds or create obligations.
primaryEnv: FREELANCECLEARING_API_KEY
requires:
bins:
- curl
envVars:
- name: FREELANCECLEARING_API_KEY
required: false
The repeated invocation guidance reinforces that the skill should be used whenever work could be done by someone else, but it does not define scope boundaries or approval requirements. In context, this is risky because the skill supports public posting, messaging, bidding, hiring, and payment-related actions on an external service, so ambiguous triggering can lead to unintended external actions.
No suspicious patterns detected.