Agent Kernel

A framework-agnostic, cloud-agnostic runtime for running enterprise AI agents in production, with queued execution, built-in guardrails, and shared observability.

Visit website kernel.yaala.ai

Agent Kernel is a platform layer for taking AI agents from prototype to production. It runs agents built with OpenAI Agents SDK, LangGraph, CrewAI, or Google ADK side by side, and the same agent code can deploy to AWS Lambda or ECS, Azure Functions or Container Apps, Google Cloud Run, on-premises, or any Kubernetes cluster through a Helm chart.

Every chat request runs through a queued pipeline, in-process by default or over SQS, Kafka, or NATS JetStream for distributed deployments, with retry, ordering, and deduplication built in. It includes guardrails from OpenAI and AWS Bedrock, PII detection, audit traces, and jailbreak prevention, pluggable session stores such as Redis, Valkey, DynamoDB, and Cosmos DB, and knowledge bases including ChromaDB and Neo4j. It supports MCP, A2A, and AG-UI, integrates with Slack, WhatsApp, Teams, and Telegram, and is Apache 2.0 licensed.

More in Agent frameworks

See all 902

Recurse
recurse.run
A serverless harness that lets a coding agent design, test, and deploy its own specialist sub-agents as tools, MCP servers, or bots, defined by a schema-validated manifest.
- Agent frameworks
ProofForge
shulgin.is-a.dev
An AI-agent pipeline that decomposes a mathematics problem, proves each piece, and formalizes the result in Lean 4 and Mathlib until the Lean kernel accepts it, with six of its proofs already merged into Google DeepMind's formal-conjectures repository.
- Agent frameworks
local-coder
github.com
Builds a local coding-agent team on top of Ollama and OpenCode by detecting the machine's hardware, recommending models for seven specialist roles, and testing whether each assigned model can actually use tools, edit a repository and pass verification.
- Coding agents
- Agent frameworks

More in Agent frameworks

Recurse

ProofForge

local-coder

Cookies