astra-campaign

Pass

Audited by Gen Agent Trust Hub on Jul 10, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill has a surface for indirect prompt injection where untrusted data provided by the user is processed by the agent.
  • Ingestion points: The agent accepts and processes user-provided product briefs, brand details, and marketing objectives as described in the SKILL.md file.
  • Boundary markers: No explicit delimiters, tags, or 'ignore instruction' warnings are used to separate the user-supplied content from the agent's core instructions.
  • Capability inventory: The agent has access to multiple high-capability tools including web_search, image_generate, video_generate, and music_generate, as well as the ability to execute its own local Python scripts (claims_check.py and validate_skill.py).
  • Sanitization: No security-specific sanitization or filtering is performed on user input; the claims_check.py script focuses exclusively on advertising compliance and regulated industry terminology.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 10, 2026, 10:39 AM
Security Audit — agent-trust-hub — astra-campaign