competitor-post-engagers
Pass
Audited by Gen Agent Trust Hub on Oct 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests and processes user-generated content from LinkedIn, which could be used to deliver malicious instructions to the agent during analysis or classification.
- Ingestion points: Scrapes profile headlines, names, and comment text from LinkedIn posts using the Apify API in
scripts/competitor_post_engagers.py. - Boundary markers: The script does not utilize specific delimiters or escape sequences when concatenating scraped content for ICP classification or exporting to CSV.
- Capability inventory: The skill can write to the local file system (CSV export) and perform network operations via the Apollo and Apify APIs.
- Sanitization: Processing is limited to basic string truncation and character replacement (e.g., removing newlines), which does not fully mitigate the risk of injection or CSV injection attacks.
- [EXTERNAL_DOWNLOADS]: The skill fetches dataset items and post engagement data from Apify.
- Evidence: The script
scripts/competitor_post_engagers.pyusesurllib.request.urlopento connect to the official Apify API (api.apify.com) to retrieve scraped content. - Context: This is a core functional requirement of the skill and uses a well-known service.
Audit Metadata