search-tcp

Pass

Audited by Gen Agent Trust Hub on Feb 26, 2026

Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
  • [SAFE]: No malicious patterns, obfuscation, or unauthorized data access were identified within the skill definition.
  • [COMMAND_EXECUTION]: The skill executes the local command npm run neta with parameters for keywords and sorting. This is consistent with its stated purpose.
  • [CREDENTIALS_UNSAFE]: The skill requires the NETA_TOKEN environment variable. This follows standard security practices for providing API tokens to CLI tools and avoids hardcoding secrets.
  • [PROMPT_INJECTION]: The skill processes user input through the --keywords parameter. While this creates a potential surface for indirect prompt injection, it is considered a low-risk architectural characteristic rather than a vulnerability in this context. Evidence: 1. Ingestion points: --keywords argument in SKILL.md. 2. Boundary markers: None. 3. Capability inventory: Subprocess execution via npm run. 4. Sanitization: Not specified in skill description.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 26, 2026, 04:24 AM