linkedin-commenter-extractor

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

• Third-party content exposure detected (high risk: 0.90). The skill explicitly calls the Apify actor "harvestapi~linkedin-post-comments" (see SKILL.md and scripts/extract_commenters.py) to fetch and ingest public, user-generated LinkedIn post comments and headlines, which the agent parses and uses (e.g., comment_text, parsed titles/companies, linkedin_url/deduping) and therefore untrusted third-party content can influence subsequent outputs and decisions.

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

• Potentially malicious external URL detected (high risk: 0.90). The skill makes runtime API calls that start an Apify actor (e.g., POST to https://api.apify.com/v2/acts/harvestapi~linkedin-post-comments/runs, or when proxied via Gooseworks: https://api.gooseworks.ai/v1/proxy/apify/acts/harvestapi~linkedin-post-comments/runs), which executes remote actor code the skill relies on—i.e., it triggers remote code execution at runtime.