skills/obra/dotfiles/syncing-obsidian/Gen Agent Trust Hub

syncing-obsidian

Pass

Audited by Gen Agent Trust Hub on Jul 7, 2026

Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill utilizes the obsync command-line tool for operations such as login, manual synchronization (pull/push), real-time monitoring (watch), and status checks.
  • [COMMAND_EXECUTION]: Provides specific instructions for using sqlite3 to manually interact with the .obsync/state.db database for troubleshooting purposes, such as resetting sync hashes and version cursors.
  • [PROMPT_INJECTION]: Surfaces an indirect prompt injection risk as the skill is designed to read and process external data from synchronization logs and vault files.
  • Ingestion points: Accessing content within .obsync/sync.log, querying .obsync/state.db, and reading user-provided files within the vault.
  • Boundary markers: No explicit delimiters or boundary warnings are provided for the agent when ingesting vault content or log data.
  • Capability inventory: Full access to shell command execution through the obsync and sqlite3 binaries.
  • Sanitization: No instructions for sanitizing file content or escaping file paths before interpolation into shell commands are documented in the skill.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 7, 2026, 04:25 AM
Security Audit — agent-trust-hub — syncing-obsidian