investor-company-prospecting

MEDIUM W011: Third-party content exposure detected (indirect prompt injection risk).

• Third-party content exposure detected (high risk: 0.90). The SKILL.md explicitly instructs the agent to fetch and parse public VC portfolio pages (e.g., https://www.ycombinator.com/companies) and to use WebSearch/call_ai to read companies' careers pages and job boards—untrusted public web content that the agent must interpret to decide which companies/contacts to target and next actions, enabling indirect prompt injection.

MEDIUM W012: Unverifiable external dependency detected (runtime URL that controls agent).

• Potentially malicious external URL detected (high risk: 0.90). The skill instructs users to run curl -s "https://code.deepline.com/api/v2/cli/install" | bash, which fetches and executes remote install code during setup and is effectively required for running the cli-based runtime commands in the skill (https://code.deepline.com/api/v2/cli/install).