li-everyday-story-persona

Warn

Audited by Gen Agent Trust Hub on Jul 13, 2026

Risk Level: MEDIUMDATA_EXFILTRATIONCOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [DATA_EXFILTRATION]: Access to personal data in the user's home directory. The skill is designed to read and write to ~/.li/个人故事库.md (Personal Story Database). While this is intended for personalizing scripts, accessing hidden files in the home directory (~/) is a sensitive operation that can expose private user history.
  • [COMMAND_EXECUTION]: The execution logic requires the agent to perform file system operations (read/write/directory creation) to manage the story database. This involves shell command execution to interact with the hidden path ~/.li/.
  • [PROMPT_INJECTION]: Potential for Indirect Prompt Injection via the local file system. The skill reads data from ~/.li/个人故事库.md and incorporates it directly into the prompt context for script generation.
  • Ingestion points: ~/.li/个人故事库.md (Step 1 of execution logic).
  • Boundary markers: Absent; the content is read and evaluated for material sufficiency without delimiters or instructions to ignore embedded commands.
  • Capability inventory: Subprocess/Shell calls for file read/write, and content generation capabilities.
  • Sanitization: None; the skill does not specify any validation or filtering of the content retrieved from the local story database.
Audit Metadata
Risk Level
MEDIUM
Analyzed
Jul 13, 2026, 10:38 AM
Security Audit — agent-trust-hub — li-everyday-story-persona