HN Showcase
Back to browse
GitHub Repository

Audit log + guard for AI agents. Passive logging, human-in-the-loop approval for dangerous ops (rm, drop, transfer) via Telegram. Diary, daily digest, timeline UI. Cursor & MCP ready. Cloudflare Workers + Hono + D1.

10 starsTypeScript

ClawDiary – Audit log and human approval gate for AI agents (Claw, MCP

by jetywolf·Mar 9, 2026·2 points·0 comments
Visit ProjectView on HN

AI Analysis

●●●BangerBig BrainShip It

Human-in-the-loop approval for AI agents via Telegram before risky ops.

Strengths
  • •Telegram approval flow is frictionless compared to custom dashboard UIs.
  • •Async logging to D1 ensures zero latency impact on agent execution.
Weaknesses
  • •Modifying agent rules (.cursor/rules) might break with future upstream updates.
  • •Self-hosting setup on Cloudflare Workers may deter non-technical users entirely.
Category
Security
Target Audience

Developers using AI coding agents

Similar To

Lakera · Guardrails AI

Post Description

ClawDiary is an audit diary and guard for AI agents. I use Cursor a lot and wanted something that (a) logs what the agent does and (b) blocks dangerous actions until I approve them in Telegram.

What it does: - Audit: agents POST to /v1/audit after each action; writes go to D1 async so the agent isn’t blocked. - Guard: before risky ops (rm, drop table, execute_bash, transfer, etc.) the agent calls /v1/guard. Request is stored and the call blocks until you approve or reject in Telegram (or by updating DB). Green-light actions return immediately. - Diary, daily digest (cron), and a minimal timeline UI so I can see what happened.

Stack: Cloudflare Worker, Hono, D1, Durable Objects (for holding the guard request until approval). Cursor integration is via a .cursor/rules rule that calls the API; MCP-capable agents can use the tool described at /mcp.json.

Repo: https://github.com/jetywolf/claw-diary Live API/docs: https://api.clawdiary.org/docs

Happy to answer questions or take feedback.

Similar Projects

AI/ML●●Solid

Axon – Agentic AI with mandatory user approval and audit logging

Agent approval gates and audit logs beat open-source alternatives, but multi-agent governance isn't novel.

Solve My ProblemBig Brain
NeuroVexon
123mo ago
AI/ML●●Solid

Axon – Open-source agentic AI with approval gates (Apache 2.0)

AXON's core move — surfacing every tool call with a low/medium/high risk label and requiring Allow/Reject/Allow-for-session — is a practical, under-explored control for agentic AI and immediately useful for regulated environments. The repo pairs that UX with a React UI, FastAPI backend, Docker sandboxed code execution and multi‑LLM integrations (Ollama/Claude/OpenAI), so it feels like a real starter stack rather than a sketch. Biggest unknowns are adoption and ecosystem of plugins/skills — the idea is solid, but it needs community momentum to matter.

Niche GemSolve My Problem
NeuroVexon
123mo ago