Asymptote-Labs

Asymptote-Labs / agent-beacon

Public

Beacon is the world’s first open-source endpoint telemetry layer for local AI agents.

docs.asymptotelabs.aicli
28
0
100% credibility
Found May 15, 2026 at 30 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Go
AI Summary

Beacon Endpoint Agent is a local monitoring tool for AI coding assistants that logs their file changes, commands, and sessions into secure, privacy-focused records viewable via a simple dashboard.

How It Works

1
🔍 Discover Beacon

You hear about a helpful tool that watches AI coding assistants on your computer to keep track of their actions safely.

2
📥 Install easily

Add the free program to your Mac with a quick download or one-click setup.

3
⚙️ Start monitoring

Turn on tracking for your AI tools like Claude or Cursor so it records what they do.

4
✨ See AI activity

Your first events appear, showing exactly what your AI tools touched or changed.

5
📊 Open dashboard

Launch a private screen to browse, search, and review all your AI sessions.

6
🔒 Secure your logs

Save records in a safe format you control, ready for review or security tools.

✅ Feel in control

You now safely see and understand every AI action on your computer.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 30 to 28 stars Sign Up Free
Repurpose This Repo

Repurpose is a Pro feature

Generate ready-to-use prompts for X threads, LinkedIn posts, blog posts, YouTube scripts, and more -- with full repo context baked in.

Unlock Repurpose
AI-Generated Review

What is agent-beacon?

Beacon is an open-source Go CLI that instruments local AI agents like Cursor, Claude Code, Codex CLI, and Claude Cowork for endpoint telemetry. It auto-configures OTLP exporters, installs Cursor hooks for session and tool events, and pipes everything into Wazuh-compatible JSONL logs—no account or cloud required. Run `beacon endpoint install` for a local collector and dashboard to query prompts, commands, file edits, and approvals.

Why is it gaining traction?

In a sea of github beacon probe or beacon simulator github repos, Beacon stands out as the first local-first layer for observing AI agent actions on laptops, normalizing OTLP traces/logs into searchable events. Devs love the zero-setup Homebrew install, macOS packaging, and Wazuh integration for existing SIEM pipelines, plus ngrok helpers for Claude Cowork testing. Its visibility-first thesis hooks teams watching agents touch files and shells without vendor telemetry.

Who should use this?

Security engineers at startups governing Cursor/Claude usage, AI devs debugging tool calls and MCP activity, or macOS teams feeding agent logs into Wazuh for compliance. Ideal for enterprises needing prompt/command audits before full agent governance.

Verdict

Grab it if you're on macOS experimenting with local agents—solid CLI, tests, and dashboard make it usable now despite 28 stars and 1.0% credibility score. Still early (watch for Linux/Windows polish), but a smart bet for the agent telemetry gap. (198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.