plaeng-rang
Pass
Audited by Gen Agent Trust Hub on Sep 10, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute a bash script to search for and list persona files within the user's local directories, including paths inside '$HOME/.claude/'. This involves automated execution of 'bash', 'ls', and 'sed' to process file system data.
- [INDIRECT_PROMPT_INJECTION]: The skill ingests user-provided text as a 'problem' and processes it within a reasoning template, creating a surface for potential instruction confusion or prompt injection.
- Ingestion points: User-supplied input for the '' argument passed to the skill.
- Boundary markers: None identified; user input is directly included in the markdown headers and reasoning sections without delimiters.
- Capability inventory: The skill uses shell execution capabilities to search local directories and list files.
- Sanitization: There is no evidence of input validation, filtering, or escaping for the user-supplied data before it is used to generate the agent's reasoning path.
Audit Metadata