finding-twitter-spaces-hosts-by-topic

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONDYNAMIC_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill documentation includes examples of bash commands to execute a local script run_actor.js and use curl for API interactions. These commands are used to trigger actors and fetch results from Apify.
  • [DYNAMIC_EXECUTION]: The input customMapFunction allows for a JavaScript function string to be passed to the scraper for data transformation, which is executed dynamically in the remote environment.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted content from Twitter, making it a surface for indirect prompt injection attacks.
  • Ingestion points: Data is fetched from Twitter via the apidojo~tweet-scraper actor as described in the Workflow section.
  • Boundary markers: There are no explicit instructions or delimiters provided to ensure the agent ignores instructions embedded within the scraped tweets.
  • Capability inventory: The skill utilizes local shell commands (node, curl) and performs network communication with the Apify API.
  • Sanitization: The process does not explicitly define sanitization or filtering logic for the scraped text before scoring or output generation.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 06:41 PM
Security Audit — agent-trust-hub — finding-twitter-spaces-hosts-by-topic