MetapriseAI

MetapriseAI / OrgKernel

Public

Open-source trust layer for AI agents — cryptographic agent identity (Ed25519), instance-scoped execution tokens, SHA-256 hash-chained audit logging, and enterprise SSO/SCIM federation. The security foundation powering every agent in the Metaprise AURA platform.

18
2
100% credibility
Found Mar 25, 2026 at 18 stars -- GitGems finds repos before they trend. Get early access to the next one.
Sign Up Free
AI Analysis
Python
AI Summary

OrgKernel is an open-source toolkit that provides secure identities, permission limits, and unchangeable activity logs for AI agents in businesses.

How It Works

1
📰 Discover secure AI management

You hear about OrgKernel, a tool that keeps company AI helpers trustworthy and tracks everything they do safely.

2
🏢 Set up for your company

You prepare it by sharing basic details about your organization so it knows who's in charge.

3
🆔 Give AI a trusted badge

You create a special company badge for your first AI helper, proving it's official and safe to use.

4
📋 Plan a job with limits

You outline a task for the AI, setting clear boundaries on what tools it can touch and how much.

5
🚀 Launch the AI job

You start the task, and the AI works only within your safe rules while noting every action.

6
📊 Review the full story

You check the unbreakable record of what happened, confirming nothing sneaky occurred.

✅ AI runs safely forever

Your company now has reliable AI helpers that stay accountable, giving you peace of mind.

Sign up to see the full architecture

5 more

Sign Up Free

Star Growth

See how this repo grew from 18 to 18 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 OrgKernel?

OrgKernel is an open source trust layer for AI agents, delivering cryptographic identities via Ed25519 keys, mission-scoped execution tokens with tool allowlists and param bounds, and SHA-256 hash-chained audit logs stored in PostgreSQL. Built in Python with FastAPI for REST APIs, it lets you bootstrap secure agent workflows—pip install, init with your org ID and SSO provider like Okta, then create identities, mint tokens, and verify audit chains via simple SDK calls or endpoints. It solves the core problem of making AI agents enterprise-ready without trusting black-box promises, enforcing zero trust open source tools for verifiable actions.

Why is it gaining traction?

In a world of hype around AI agents, OrgKernel stands out with production-grade crypto primitives like challenge-response auth and tamper-evident logs, plus 100% test coverage across 61 cases—no half-baked experiments here. Developers dig the no-vendor-lock-in Apache 2.0 license, self-hosted FastAPI integration, and hooks for open source trusted execution environments, making it a practical alternative to proprietary agent security stacks. Early adopters get tamper-proof compliance trails out of the box, with SSO/SCIM federation for real orgs.

Who should use this?

Security engineers at AI startups building multi-agent platforms need it for audit-compliant tool calls and identity revocation. Compliance teams in finance or healthcare evaluating agent orchestration will value the policy engine and authority graphs for L0-L5 approvals. Devs prototyping zero trust agent networks on GitHub open source tools can drop it in via pyproject.toml for instant PKI and token gating.

Verdict

Grab it if you're serious about secure AI agents—solid docs, full tests, and pip-ready make it usable now despite 18 stars and 1.0% credibility score signaling early maturity. Fork and harden for prod; it's a strong Phase 1 foundation worth watching.

(198 words)

Sign up to read the full AI review Sign Up Free

Similar repos coming soon.