Agent infrastructure
harbor
harbor-framework/harbor
A framework for evaluating and improving agents, supporting parallel multi-agent benchmarks, custom environments, and training-data generation for reinforcement learning workflows.
Topics
Related projects
agent-skills-platform2.4kA platform for building tested agent skills and governing their lifecycle through a team marketplace, with approvals, versioned releases, rollback, quarantine, and multi-platform distribution.- agent-skill-creator2.4kA skill-creation workflow that turns real work into tested, installable agent skills with evals and governed team distribution.
waku-agent2.2kA local-first personal AI assistant with its own agent loop, layered memory, evals and chat gateways, shipped as readable Python code with all state in a local SQLite file.
paperthin1.4kLow-level agentic design patterns. Turning old engineering wisdom into reflexes your agent reaches for on its own-on any agent.
cc-switch152.6kA cross-platform desktop manager for coding agents that switches API providers and centralizes MCP, skills, and prompts.
OmniRoute86.2kA free MIT AI gateway that exposes hundreds of providers and models through one OpenAI-compatible endpoint, with quota-aware fallback, token compression, routing strategies and MCP/A2A support.