blog-post-writer

Pass

Audited by Gen Agent Trust Hub on Apr 5, 2026

Risk Level: SAFE
Full Analysis
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill retrieves the SEEDREAM_API_KEY from environment variables, which is a secure practice for managing API credentials.
  • [DATA_EXPOSURE_AND_EXFILTRATION]: The skill performs network requests to api.gptnb.ai to generate images. This external communication is disclosed and aligned with the tool's core functionality.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted user input, including raw text and reference links, which could potentially contain malicious instructions.
  • Ingestion points: Scattered thoughts, original articles, and reference links provided by the user as described in SKILL.md.
  • Boundary markers: Absent; the skill processes user input directly without specific delimiters or instructions to ignore embedded commands.
  • Capability inventory: The skill calls the generate_image tool and writes output to a new Markdown file on the local file system.
  • Sanitization: No sanitization, validation, or filtering of the user-provided content is implemented.
Audit Metadata
Risk Level
SAFE
Analyzed
Apr 5, 2026, 07:35 AM
Security Audit — agent-trust-hub — blog-post-writer