10-andruia-skill-smith

Warn

Audited by Gen Agent Trust Hub on Jul 24, 2026

Risk Level: MEDIUMCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill instructs the agent to perform file system operations, including creating folders and writing files to a hardcoded local path (D:\...\agentic-awesome-skills\skills\). This direct interaction with the host file system should be limited to trusted environments.
  • [PROMPT_INJECTION]: The skill contains an indirect prompt injection surface where untrusted user input is used to generate the logic and roles of new agent skills.
  • Ingestion points: User-supplied input for 'Nombre Técnico', 'Rol Experto', and 'Outputs Clave' during the 'ADN de la Skill' phase (SKILL.md).
  • Boundary markers: Absent. The instructions do not utilize delimiters or 'ignore' directives to prevent the agent from following instructions embedded within the user's provided data.
  • Capability inventory: The skill performs file-writing operations and repository updates based on generated content (SKILL.md).
  • Sanitization: Absent. There is no evidence of validation, escaping, or filtering applied to the user input before it is written to the file system.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 24, 2026, 06:18 PM
Security Audit — agent-trust-hub — 10-andruia-skill-smith