Symbiont: AI-Native Agent Framework

What

Symbiont Is Symbiont is an AI-native agent framework for building autonomous, policy-aware agents. It targets agent builders, AI developers, and platform engineers, offering scheduling, channel adapters, cryptographic identity, and multi-language SDKs.

Core Capabilities

• Agent lifecycle management: Create, start, stop, pause, resume, and destroy agents.
• Agent management API: List agents, get status, create, update, delete, execute, get execution history, heartbeat, and push events to agents.
• Scheduling: Create schedules.
• Channel adapters: Integrate agents with chat platforms.
• Cryptographic operations: Native support for encryption, signing, and proofs.
• Multi-language SDKs: Integrate across different technology stacks.
• AI assistant plugins: Governed Claude Code subprocess and Gemini CLI.
• Conditional compilation, macros, and code generation: Advanced features and integration with external systems.

Architecture

and Boundaries The branch retains ORGA, ToolClad, Cedar, SchemaPin, and AgentPin. Default reason()/tool_call() use a frozen project root and protected journals. A reasoning loop stops with UnconfirmedEffects after a dispatched tool error, unconfirmed nested network outcome, or started child that does not complete. A rejected module retains existing agents and helpers.

Configuration

and Security Ambiguous agent selection, conflicting settings, and registered HTTP security-tier mismatches are configuration errors. Creating an agent does not adopt its permissions. An automatically launched supervisor survives its caller's death, but a shared service manager or outer container can terminate both processes together. Broader workspace/platform trust review remains open.