Hermes Agent
A persistent personal agent with memory, reusable skills, tools and messaging gateways.
Explore 18 tools and 11 practical workflows for coding, research, videos, games and everyday work.
November 2026 edition · Curated by RohitAI, with features, costs and official sources.
Repository versions, licences and status · Mondays, 08:10 IST
What is an agent harness? The software around a model that manages its tools, context, permissions and task loop—so it can act, check results and keep working.
Choose the core agent loop. Harnesses, SDKs and lower-level frameworks are labeled separately.
A persistent personal agent with memory, reusable skills, tools and messaging gateways.
A coding agent with a terminal interface, a desktop app in beta, and separate build and planning modes.
A minimal coding-agent toolkit designed to be extended through packages, tools and user-defined workflows.
An opinionated agent harness with planning, filesystem tools, context management and configurable subagents.
Anthropic’s agentic coding tool for repository work, terminal commands and tool integrations.
OpenAI’s coding harness for repository work, with a CLI, SDKs and an app-server interface for custom products.
An extensible local agent that can work with code and connect to external tools through MCP extensions.
A Pi-derived terminal coding agent with language-server tools, debugging support and a configurable development workflow.
Google’s terminal agent maintained for enterprise Gemini Code Assist and supported paid API workflows. Individual Google AI plans moved to Antigravity CLI.
A coding agent available through editor integrations, a CLI, a desktop app and an SDK, with tools, checkpoints and configurable approvals.
A coding-agent toolkit offering editor and terminal workflows with access to multiple model providers.
Agent Canvas provides an interface for OpenHands agents. Its separate SDK, agent server and automation service supply execution and scheduling.
A terminal coding assistant that edits an existing codebase and integrates changes with Git.
A browser-agent library with local execution, a CLI and optional hosted agents and browser infrastructure.
A deliberately minimal software-engineering agent designed to make the core agent loop easy to understand.
A framework for constructing stateful agent workflows with explicit control over execution and persistence.
An SDK for custom agent applications with tools, handoffs, approvals and tracing, plus sandbox agents for file and command workflows.
A personal AI-assistant platform connecting models, tools, channels and scheduled workflows.
We use current community discussions to find useful workflows, then check official documentation, repository ownership, licenses and release status. Every profile links its evidence.
Repository data and editorial recommendations are checked separately. If a source check fails, the last verified collection stays available.
No. The model does the reasoning; the harness supplies the execution loop, tools, context management and permission boundaries around it. Different harnesses can use the same model.
Yes. This library separates a core harness from a composed workflow. Claude Code plus Remotion and its skills is a video stack; Remotion by itself is a renderer, not an agent harness.