building-twitter-prospect-lists

Pass

Audited by Gen Agent Trust Hub on Sep 2, 2026

Risk Level: SAFE
Full Analysis
  • [EXTERNAL_DOWNLOADS]: The skill makes requests to Apify's official API (api.apify.com) to run scrapers. This is the core functionality and uses standard, well-documented endpoints for the vendor's service.
  • [COMMAND_EXECUTION]: The skill provides instructions for running a local script scripts/run_actor.js and using curl to interact with the Apify API. These commands are transparent, document their purpose clearly, and are standard for developer-oriented tools.
  • [CREDENTIALS_UNSAFE]: The skill correctly instructs users to manage their APIFY_TOKEN via environment variables or .env files, which is a recommended security best practice for secret management.
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests data from Twitter (bios and tweets). While this is an ingestion of untrusted data, the skill uses it for structured filtering and display (Step 4 and 5) rather than directly interpolating it into executable logic or critical system prompts. The risk is minimized by the analytical nature of the workflow.
  • [DYNAMIC_EXECUTION]: One input parameter, customMapFunction, allows for a JavaScript string to be passed to the Apify actor. This execution happens within the isolated Apify environment, not on the local agent's system, which is a standard feature of the targeted platform.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 2, 2026, 07:16 PM
Security Audit — agent-trust-hub — building-twitter-prospect-lists