finding-freelancers-by-skill-on-twitter

Warn

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: MEDIUMDYNAMIC_EXECUTIONINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [DYNAMIC_EXECUTION]: The customMapFunction input parameter accepts a string described as a "JavaScript function to transform each output object." If evaluated at runtime, this allows for dynamic code execution based on input strings.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from Twitter (bios and tweets) to identify and score candidates.
  • Ingestion points: Twitter profiles and tweet content (SKILL.md).
  • Boundary markers: No explicit markers are defined to isolate untrusted Twitter content from agent instructions.
  • Capability inventory: Subprocess execution of Node.js scripts (scripts/run_actor.js) and network operations via curl to the Apify API.
  • Sanitization: No sanitization or filtering logic is provided for the retrieved social media content before it is scored or displayed.
  • [COMMAND_EXECUTION]: The skill directs the agent to execute shell commands, including node scripts/run_actor.js and curl for interacting with the Apify REST API. These operations are core to the skill's data collection functionality.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Sep 2, 2026, 06:36 PM
Security Audit — agent-trust-hub — finding-freelancers-by-skill-on-twitter