finding-speaking-opportunities-on-twitter

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONDATA_EXFILTRATIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides instructions to execute local Node.js scripts (scripts/run_actor.js) and shell commands (curl) to interact with the Apify API. These commands are used to trigger data scraping and fetch results.
  • [DATA_EXFILTRATION]: The skill requires and uses an APIFY_TOKEN environment variable to authenticate with external API endpoints (api.apify.com). While it involves sending search parameters to a third-party service, this is a core functional requirement of the scraper.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Twitter (tweet content).
  • Ingestion points: External data enters the context via the tweet-scraper results in Step 2 and Step 3.
  • Boundary markers: The skill lacks explicit instructions to ignore instructions embedded within the scraped tweets.
  • Capability inventory: The skill includes file writing (CSV/JSON output), command execution (Node.js/curl), and network operations (Apify API).
  • Sanitization: There is no mention of sanitizing or escaping the tweet content before it is processed by the agent or written to files.
  • [DYNAMIC_EXECUTION]: The customMapFunction input parameter allows for a JavaScript function string to be passed to the scraper for data transformation. While executed within the Apify platform's environment rather than the local agent directly, it represents a dynamic code execution surface.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 06:24 PM
Security Audit — agent-trust-hub — finding-speaking-opportunities-on-twitter