apify-ads-intelligence

Pass

Audited by Gen Agent Trust Hub on Aug 13, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill facilitates the execution of Apify Actors using the apify-cli. It instructs the agent to construct and run shell commands such as apify actors call and apify datasets get-items. These commands include dynamic parameters, Actor IDs, and JSON input blocks derived from user intents. While these are standard operations for the Apify platform, shell command assembly requires careful handling of metacharacters and input sanitization to prevent potential command injection.
  • [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection as it retrieves and processes untrusted content (ad copy, headlines, landing page descriptions) from public ad libraries (Meta, Google, TikTok, LinkedIn, X).
  • Ingestion points: External ad data enters the agent's context through the apify datasets get-items command during the synthesis phase (Step 5).
  • Boundary markers: The skill does not define explicit boundary markers or "ignore embedded instructions" warnings when presenting the fetched ad data to the agent for synthesis.
  • Capability inventory: The agent has the capability to execute shell commands (apify-cli) and perform network operations via the CLI.
  • Sanitization: There is no evidence of sanitization or filtering of the retrieved ad content before it is interpolated into the synthesis prompt.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 13, 2026, 04:14 PM
Security Audit — agent-trust-hub — apify-ads-intelligence