Hermes Agent
BetaSelf-improving learning loop. Model-agnostic by design.
Open-source autonomous agent framework built around a self-evaluation loop — creates skills from experience, refines them through continued use, and builds a persistent user model across sessions. Model-agnostic with native support for Nous Portal (Hermes 4), MiniMax, Kimi/Moonshot, GLM, OpenAI, Anthropic, and any OpenAI-compatible endpoint. Signal support not present in OpenClaw.
- Category
- Personal agents
- Runtime
- Python
- Pattern
- SDK / Library
- Adapter package
- intramedial-actor-hermes
Status
Hermes Agent is Beta.
The integration path is built and tested. Customers can attach now; expect minor breaking changes until the 1.0 contract freeze.
Ships from the repo as `intramedial-actor-hermes` — not yet on a package registry.
How to attach
SDK / Library.
Install the adapter
Add intramedial-actor-hermes to your project.
Issue a workspace API key
Generate a key in the dashboard under Settings → API Keys, scoped to the workspace and environment this agent's data should land in.
Record each run
Construct the adapter once with your token and workspace id where you initialise your Hermes Agent agent, then call recordRun(run) when a run finishes. The adapter maps the run to signed receipts and, by default, captures a failed attestation instead of crashing your agent.
Verify
Trigger one agent run. The Witness page shows the record once the agent has emitted it.
What gets recorded
4 event types, captured today.
- agent.invocation
- Every top-level agent run — the input, the resolved output, timing and token usage.
- tool.call
- Each tool or function the agent invokes, with its arguments and returned result.
- memory.write
- Writes to the agent's persistent memory store, so state changes stay auditable.
- channel.message
- Messages sent and received over the agent's channels (Telegram, Slack, Discord, …).
Same category
Others in Personal agents.
- OpenClawGA
The original personal-agent framework. First-class integration.
- NanoClawBeta
Container-isolated, security-first. Built on the Claude Agent SDK.
- ZeroClawRoadmap
Rust-native. A single binary you configure and run.
- NanobotBeta
~4,000 lines of code. Built for MCP-first agents.
- NemoClawBeta
NVIDIA's sandboxed reference stack for running agents.
- IronClawRoadmap
NEAR AI's Rust agent OS. Privacy and security first.
Start recording Hermes Agent agents.
Every Hermes Agent run lands in the same canonical Witness record as every other framework in the catalogue: signed, replayable, and exportable as evidence.
A Trustvice ProductAgents act when the plane says yes, and the record says why.
Put admission in front of your MCP servers, repositories, SaaS tools and cloud accounts. Every agent gets a lease or a reason – and you keep the lever to stop any of them.
