Your data stays
on your machine.

• OS Keyring Credentials — Passwords, passphrases, and private keys are stored in your operating system's secure credential store — never written to disk or local storage in plaintext.
• Credential Stripping — When connection profiles are saved, sensitive fields are automatically stripped. Only non-sensitive metadata is persisted.
• Read-only AI Guardrails — Every AI-authored query is validated to ensure only SELECT, SHOW, and EXPLAIN are permitted. DROP, DELETE, and TRUNCATE are blocked before execution — per database dialect.
• Query LIMIT Enforcement — Result limits are automatically applied to prevent accidental large fetches, while respecting any existing limits you've set.
• PII Scanner — Query results are scanned for sensitive data patterns including SSNs, credit cards, emails, phone numbers, and IBANs. Redacted previews keep sensitive data out of AI output.
• Enterprise Privacy Mode — When privacy mode is enabled, the AI cannot execute queries against your data. It reasons from schema metadata only and authors SQL for you to run — zero data egress.
• Audit Log — Every AI run is logged locally with mode, target, tool calls, guardrail blocks, rows scanned, and outcome — available for your review anytime.
• Local-First by Default — The app runs entirely on-device. Schema metadata, semantic notes, and search indexes stay local. You choose whether to connect a cloud AI provider.

Beyond SAYA AI requests you initiate yourself, the app makes only two other kinds of network calls: to our license server (when starting a trial or activating/deactivating a license) and a periodic check of a static update manifest hosted on GitHub Releases. There is no telemetry or usage analytics of any kind.

Two privacy modes

Full audit trail

Every SAYA AI run — Quick Chat, Skill, or Goal — is logged locally. You can inspect it anytime: