monitoring-brand-mentions-on-twitter

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTIONEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Twitter/X search results (tweet text, author handles, and URLs) and processes it for sentiment analysis and reporting.
  • Ingestion points: Fetches data from Twitter via Apify actors (apidojo/tweet-scraper) in Step 2 and Step 3.
  • Boundary markers: None identified in the instructions for isolating external tweet content.
  • Capability inventory: The skill uses curl for network requests and node to execute local scripts (run_actor.js).
  • Sanitization: No specific sanitization or filtering of tweet content is mentioned before it is processed by the agent or output in the final report.
  • [DYNAMIC_EXECUTION]: The customMapFunction input parameter accepts a raw JavaScript function string intended to transform output objects. While this is a common feature in data processing tools, it represents a dynamic execution surface if the agent allows the execution of this user-supplied code.
  • [EXTERNAL_DOWNLOADS]: The skill makes network requests to api.apify.com to run scrapers and retrieve datasets. These requests use the APIFY_TOKEN provided by the user.
  • [COMMAND_EXECUTION]: The instructions direct the agent to execute shell commands using node and curl to interact with the Apify API and local scripts. These commands include environment variable usage ($APIFY_TOKEN, $RUN_ID) and parameter interpolation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 06:44 PM
Security Audit — agent-trust-hub — monitoring-brand-mentions-on-twitter