tracking-brand-sentiment-across-platforms

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill makes network requests to api.apify.com to trigger data collection actors. This is a well-known service and the behavior is expected for the skill's stated purpose of social media scraping.
  • [COMMAND_EXECUTION]: The skill instructs the agent to execute curl commands for API interaction and a local Node.js script (scripts/run_actor.js) to process data and export results to CSV or JSON files.
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest and analyze untrusted data from Twitter, Reddit, and TikTok, creating an attack surface for indirect prompt injection.
  • Ingestion points: Raw posts and comments are fetched from social media platforms via external scrapers defined in the workflow.
  • Boundary markers: The instructions do not specify any delimiters or warnings to the agent to disregard instructions embedded within the social media posts.
  • Capability inventory: The agent has access to shell commands (curl) and the ability to execute local scripts that interact with the file system for data storage.
  • Sanitization: The workflow lacks a verification or sanitization step to filter malicious instructions or malformed data from the scraped social media content before analysis.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 06:58 PM
Security Audit — agent-trust-hub — tracking-brand-sentiment-across-platforms