setup

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: Uses local shell commands to manage configuration files and verify authentication status. It specifically creates a local directory (~/.claude/overleaf-skills) and writes the session cookie to a file with restricted permissions.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a surface for processing user-supplied data (the cookie value). It includes explicit validation logic to normalize the input and reject potentially malicious content containing newlines or quotes, which prevents command injection during the file-writing process.
  • [CREDENTIALS_UNSAFE]: Manages the collection of a sensitive authentication token (overleaf_session2 cookie). The skill follows security best practices for credential handling by instructing the user to perform manual extraction and applying strict file system permissions (chmod 600) to the resulting storage file.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 07:56 AM