v1.2.0 - Vault Tag Matching
What's new
Vault Tag Matching
- AI-generated tags are matched against all existing vault tags before being applied
- Case-insensitive matching: vault casing always wins (e.g. AI returns
machine-learning, vault hasMachine-Learning→ applied asMachine-Learning) - Unmatched tags pass through as-is from the AI
- Two-layer approach: vault tags sent to AI as prompt context, then AI output post-processed with deterministic lookup
- New setting: Prefer Existing Vault Tags (default: enabled)
Also included (v1.1.0)
- Multi-language support: English and French
- Rate limiting: configurable requests per minute (default: 10)
- Retry logic with exponential backoff (default: 3 retries)
See updates.md for full changelog.