instar
Persistent Claude Code agents with scheduling, sessions, memory, and Telegram.
- 86
- TypeScript
- MIT
- 10 h ago
Persistent Claude Code agents with scheduling, sessions, memory, and Telegram.
Auth for AI agents and humans. First-class agent identity, MCP OAuth 2.1, delegation, audit. TypeScript, edge-native, MIT.
Open-source preview: deploy existing agent containers to local Docker behind a deploy-time identity gate (Entra, Keycloak, Okta, AWS), with Cedar/OPA policy, second-person approval, a kill switch and a hash-chained audit log.
Static browser tool to publish a compact owner-signed agent identity and capability card.
Verify and sign AI agent HTTP traffic in Python — RFC 9421 · Web Bot Auth · AAuth
Identity registry and lifecycle management for fleets of AI agents. Zero-dependency TypeScript IAM: cryptographic agent identity (Ed25519), declared capabilities with fail-closed authorization, an enforced lifecycle (provision, activate, suspend, revoke, expire, archive), A2A Signed Agent Cards, and a hash-chained audit trail.
智能体可信身份与持续核验技术要求 v0.1 · Technical Requirements for Trusted Agent Identity and Continuous Verification
An operating system for agent identity and memory — soul-bootstrap, memory-layers, soul-audit. Cross-agent SKILL.md, zero deps.
Agent SOC: a 3-minute game about AI agent security. Review agent actions by hand, with an AI supervisor, and with architecture that can't be talked round. English + Russian.
AAuth Explained: an interactive visual guide to the AAuth protocol (IETF draft) for AI agent identity and authorization. English + Russian.
Open, vendor-neutral vocabulary for how AI agents are identified, authorized, credentialed and audited: 9 dimensions, 40 terms, JSON Schema, worked examples.
Infrastructure-grade identity, authorization and audit control plane for AI agents
Quickstart for the Universal Agentic Registry — discover and connect with 46,000+ AI agents across all protocols. One API for Web2 and Web3 agents.