Pfmea
Pass
Audited by Gen Agent Trust Hub on Feb 17, 2026
Risk Level: SAFENO_CODE
Full Analysis
- [NO_CODE] (SAFE): The skill is comprised solely of Markdown files. No scripts (Python, JavaScript, Shell) or binaries are included or referenced for execution.
- [COMMAND_EXECUTION] (SAFE): There are no instances of system command execution, shell spawning, or subprocess calls.
- [DATA_EXFILTRATION] (SAFE): No network-capable utilities or code (such as curl, wget, or API requests) were found, and no data is transmitted to external domains.
- [DATA_EXPOSURE] (SAFE): The skill includes instructions to read other local skill files within the
~/.claude/skills/directory. This is used for contextual integration with related manufacturing skills and does not involve accessing sensitive system files, credentials, or private user data. - [PROMPT_INJECTION] (SAFE): The instructions are focused on manufacturing methodology and do not contain attempts to override agent safety filters or system prompts.
Audit Metadata