kontext-security

kontext-security / kontext-cli

Public

Open-source CLI for AI coding agents. Give your coding agents access to services without exposing keys.

ai ai-agents audit-logging cli credential-management
100
1
100% credibility
Found Apr 15, 2026 at 107 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Go
AI Summary

Kontext CLI is an open-source command-line tool for securely managing short-lived credentials and governance telemetry for AI coding agents like Claude Code.

How It Works

1
👋 Discover Kontext

You hear about Kontext, a helper that keeps AI coding assistants safe by managing access securely.

2
📥 Get the tool

You install it quickly using your system's package manager or a direct download.

3
🚀 Start your first session

You run the start command, it opens your browser for a quick login, and sets everything up automatically.

4
🔗 Link your services

It guides you to connect accounts like GitHub or others if they're not ready yet.

5
🤖 Launch your AI assistant

Your AI coding helper opens with temporary secure access to your services, ready to work.

6
📊 Work with oversight

As you use the AI, it tracks actions safely without storing secrets long-term.

✅ Session ends securely

When done, access expires automatically, leaving your projects safe and logged.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 107 to 100 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 kontext-cli?

Kontext CLI is a Go-built open-source CLI that equips AI coding agents like Claude Code with secure, short-lived credentials for services such as GitHub or Stripe, ditching risky long-lived API keys in .env files. Run `kontext start --agent claude` to handle OIDC login, create a local .env.kontext with placeholders, exchange them via RFC 8693 for ephemeral tokens, and launch the agent with hooks logging tool calls to a dashboard. As a github open source tool and open source CLI agent, it keeps secrets out of your repo while enabling seamless agent access.

Why is it gaining traction?

This open source CLI coding agent wrapper shines by automating credential resolution and governance in one command—no Docker, no manual setup, no key rotation hassles. Teams notice instant wins like scoped tokens that auto-expire on session end, plus telemetry on PreToolUse and PostToolUse events without capturing prompts. Its lean native binary and browser-based auth make it a frictionless upgrade over copy-paste secrets.

Who should use this?

Dev teams relying on Claude Code for repo interactions or API calls, especially those tired of key leaks in AI workflows. Security leads at startups mandating audit trails for agent actions, or backend engineers integrating Stripe/databases without exposing tokens. Perfect for open source CLI users evaluating agent governance before scaling to Cursor.

Verdict

Grab it via brew if Claude Code is your daily driver and key hygiene matters—quick to test, solid docs. At 100 stars and 1.0% credibility score, it's nascent with Claude-only support; production teams should watch for maturity before org-wide rollout.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.