benchmarking-instagram-influencer-engagement
Pass
Audited by Gen Agent Trust Hub on Sep 2, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill manages authentication through the APIFY_TOKEN environment variable, adhering to security best practices for credential handling.
- [SAFE]: Network communication is restricted to api.apify.com, which is the official endpoint for the Apify platform used to run the scraping tool.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests data from public Instagram profiles (posts and comments). While this constitutes an attack surface for indirect prompt injection, the risk is minimal as the logic is focused on quantitative data processing.
- Ingestion points: Instagram post content, comments, and profile data retrieved by the apidojo/instagram-scraper actor in SKILL.md.
- Boundary markers: None specified in the instructions for separating scraped text from processing logic.
- Capability inventory: The skill performs arithmetic calculations and generates reports; it does not invoke local shell commands or file writes based on the scraped content.
- Sanitization: No explicit sanitization or escaping of scraped text is mentioned.
- [DYNAMIC_EXECUTION]: The customMapFunction parameter allows users to provide a JavaScript function for data transformation. This code is executed within the remote Apify actor environment rather than the local agent context, following the vendor's standard operational design.
Audit Metadata