Oracle-H
Human-on-the-loop validation gate for autonomous AI agents via Telegram approval workflows.
About
Provides a mandatory human checkpoint for autonomous AI agents before executing critical, destructive, or irreversible actions. Agents submit actions for validation, and a designated human approver receives a Telegram notification with full context including action details, rationale, and projected impact. Decisions are cryptographically signed with HMAC-SHA256 and returned to the agent, which proceeds or aborts accordingly. Includes anti-hallucination watchdog, TTL-based request expiry, and bearer token authentication.
Is this your product? Claim this listing with an email at your domain to post updates.
Is this your project?
Claim this listing to manage your page, access analytics, and unlock upgrades. Verification takes 60 seconds.
Share This Project
Embed Badge
Add this badge to your README:
[](https://hifriendbot.com/ai-list/oracle-h/)
