skills/klerith/fernando-skills/spec/Gen Agent Trust Hub

spec

Pass

Audited by Gen Agent Trust Hub on May 23, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: No malicious patterns or security risks were identified. The skill is intentionally designed with a human-in-the-loop flow that prevents the agent from making autonomous decisions or writing executable code.\n- [SAFE]: The skill interacts with local project files (CLAUDE.md and the specs/ directory) to maintain consistency with existing documentation. This behavior is standard for development-oriented agents and does not involve exfiltration or access to sensitive system files like SSH keys or environment variables.\n- [PROMPT_INJECTION]: The skill consumes untrusted data from existing project files, creating an indirect prompt injection surface. However, the risk is negligible due to the absence of dangerous capabilities.\n
  • Ingestion points: The agent reads the contents of CLAUDE.md and files within the specs/ directory.\n
  • Boundary markers: Absent; the skill does not utilize specific delimiters to isolate external file content from its instructions.\n
  • Capability inventory: Limited to file listing (ls), file reading (cat), and file writing to the specs/ directory. No network access, shell command execution, or package management tools are enabled.\n
  • Sanitization: Absent; external content is directly integrated into the context for text generation.
Audit Metadata
Risk Level
SAFE
Analyzed
May 23, 2026, 03:42 PM
Security Audit — agent-trust-hub — spec