sync-github-to-obsidian

Pass

Audited by Gen Agent Trust Hub on Feb 17, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION] (SAFE): The skill utilizes standard bash utilities (ls, mkdir, find, cp) to manage local documentation files. These commands are consistent with the stated purpose of the skill.
  • [DATA_EXFILTRATION] (SAFE): No network operations or external data transfer mechanisms (like curl or wget) were identified. Data movement is restricted to the local filesystem.
  • [CREDENTIALS_UNSAFE] (SAFE): While the skill contains a hardcoded absolute path (/Users/danieltang/GitHub), this is an identity exposure (PII) rather than a credential or secret leak.
  • [PROMPT_INJECTION] (LOW): The skill exhibits an indirect prompt injection surface because it reads and processes the content of markdown files originating from external repositories.
  • Ingestion points: Any .md file located within the /Users/danieltang/GitHub directory structure.
  • Boundary markers: None are implemented; the agent processes file content directly.
  • Capability inventory: The skill can list directories, create folders, and copy files locally.
  • Sanitization: No sanitization or content filtering is performed on the files before processing.
Audit Metadata
Risk Level
SAFE
Analyzed
Feb 17, 2026, 06:41 PM