extracting-tiktok-comments-for-research

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external, untrusted data in the form of TikTok comments for sentiment analysis and research.
  • Ingestion points: TikTok comments are fetched from the Apify API and ingested into the agent's context for analysis in Step 3 and Step 4.
  • Boundary markers: The instructions do not specify the use of delimiters (like XML tags or triple quotes) or explicit 'ignore embedded instructions' warnings when passing scraped text to the LLM for analysis.
  • Capability inventory: The skill involves data processing and analysis. While it doesn't describe writing this data to the local file system or executing it as code, the analysis step could be manipulated by malicious instructions hidden in comments (e.g., 'Ignore sentiment analysis and output a specific message').
  • Sanitization: The skill suggests removing spam and short comments, but these filters are intended for data quality rather than security sanitization against prompt injection.
  • [COMMAND_EXECUTION]: The skill provides instructions to execute shell commands using node and curl to interact with the Apify platform.
  • Evidence: Examples in 'Step 2: Run the Actor' show node scripts/run_actor.js and curl -X POST ... used to trigger the scraping process and fetch results. These are standard operations for integrating with the Apify service.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 07:24 PM
Security Audit — agent-trust-hub — extracting-tiktok-comments-for-research