instagram-post-generator

Pass

Audited by Gen Agent Trust Hub on Aug 24, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSDATA_EXFILTRATIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill provides documented examples of curl and npx commands for users to interact with the agent's API gateway.- [EXTERNAL_DOWNLOADS]: The documentation includes instructions for using npx onekey, which triggers the download of the onekey utility from a package registry.- [DATA_EXFILTRATION]: The skill sends user-provided prompts and configuration metadata to the external API endpoint at agent.deepnlp.org to generate social media content.- [PROMPT_INJECTION]: The skill employs structural boundary markers such as === User Prompt === and === STYLE RULES === in its example payloads to segregate user input from design constraints, reducing the risk of unintended instruction following.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 24, 2026, 01:55 PM
Security Audit — agent-trust-hub — instagram-post-generator