obsidian-search

Pass

Audited by Gen Agent Trust Hub on Jul 29, 2026

Risk Level: SAFE
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/vault_search.py invokes ripgrep (rg) to search for text patterns within the Obsidian vault. It correctly uses a list-based argument structure with subprocess.run and applies re.escape() to user-supplied search terms, preventing shell injection vulnerabilities.
  • [EXTERNAL_DOWNLOADS]: The scripts/meili_search.py script communicates with a Meilisearch server. While it makes network requests, it targets a user-defined or local endpoint (default 127.0.0.1:7700) to perform search operations, which is consistent with the skill's intended purpose.
  • [CREDENTIALS_UNSAFE]: The skill attempts to load a Meilisearch search key from an environment variable or a local file at ~/meilisearch/search-key.txt. This is a standard practice for local tool configuration and does not involve hardcoded secrets or unsafe credential handling.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 29, 2026, 03:30 AM
Security Audit — agent-trust-hub — obsidian-search