secrets-management

Pass

Audited by Gen Agent Trust Hub on Aug 26, 2026

Risk Level: SAFE
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external files (code, config, .env) which creates a surface for indirect prompt injection.\n
  • Ingestion points: File content is ingested via Read, Grep, and Glob tools during discovery (SKILL.md).\n
  • Boundary markers: The skill includes a 'Prompt Injection Safety Notice' instructing the agent to treat all file content as untrusted and ignore embedded instructions (SKILL.md).\n
  • Capability inventory: The skill is restricted to Read, Grep, and Glob tools with no network or write permissions (Frontmatter).\n
  • Sanitization: The skill mandates that the agent never extracts or reproduces actual secret values.\n- [DATA_EXPOSURE]: The skill searches for sensitive files and credential patterns (e.g., .env, .pem, AWS keys) as part of its auditing function.\n
  • Discovery patterns: It identifies paths for environment variables (.env), private keys (.pem), and cloud credentials (.aws/credentials).\n
  • Mitigation: The skill's primary purpose is security analysis and it is governed by explicit non-exfiltration safety guidelines that forbid logging or displaying actual secret values.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 26, 2026, 01:49 PM
Security Audit — agent-trust-hub — secrets-management