Building agents should be easy

Herm provides secure infrastructure for customer-facing and internal agents: MCP, memory, tools, skills, sandboxes, permissions, and observability out of the box.

Get an API key

Quick start with an agent

Cursor

Codex

Claude Code

Openclaw

Hermes

01 / PRODUCT INTERFACE

Deploy AI agents with one API call.

Stop harness-engineering. Bring your system prompt, skills, tools, and connectors, then get infrastructure, orchestration, memory, self learning, and more out of the box. Then connect over SSE or use directly in Slack.

Create an onboarding automation for new customers.

create-agent.http

1# Create an agent with the REST API
2POST https://api.herm.run/v1/agents HTTP/1.1
3Content-Type: application/json
4x-api-key: $PRISM_API_KEY
5
6{
7  "name": "Acme Marketing Agent",
8  "modelName": "anthropic/claude-sonnet-4.5",
9  "systemPrompt": "You are Acme's marketing agent.",
10  "skills": [],
11  "tools": [],
12  "mcpServers": []
13}
14
15
16
17
18
19
20
21
22
23

Preview

api.herm.run/v1/agents/agent_7xK9s2

Agent created

agent_7xK9s2

Status: active
Model: claude-sonnet-4.5
Version: 0
API: REST

Agents

Reusable versioned configuration

Read docs

Sessions

Durable threads and runs

Read docs

Streaming

Tokens and tool calls over SSE

Read docs

Observability

Usage and events for your logs

Learns every user

Scheduled background runs

Read docs

Skills

Markdown playbooks on demand

Read docs

Tools

Web, files, and shell built in

Read docs

MCP

Your tools, open protocol

Read docs

Slack

DMs and channel mentions

Read docs

Files

A persistent workspace

Read docs

Secrets

Credentials, never in the sandbox

Read docs

Ralph Wiggim loop

Send /goal and keep going

Read docs

Sandboxes

Isolated per agent and subject

Read docs

Self learning

Improves between sessions

Read docs

Access control

Gate tool calls by policy

Read docs

02 / HARNESS ENGINEERING

Build your product, not another harness.

A custom harness turns sessions, memory, files, sandboxes, credentials, automations, and streaming into your roadmap. Herm keeps that plumbing off your plate so your team can focus on the prompt, data, tools, and UX that make the agent worth using.

CapabilityHermInternal

Durable sessionsBUILT INBUILD + OPERATE

Memory across sessionsBUILT INBUILD + TUNE

Per-customer isolationBUILT INBUILD + SECURE

SSE event streamingBUILT INBUILD + SCALE

Persistent filesystemsBUILT INBUILD + MOUNT

MCP servers and toolsBUILT INBUILD + SCOPE

Scoped credentialsBUILT INBUILD + ROTATE

Scheduled automationsBUILT INBUILD + MONITOR

Human approvalBUILT INBUILD + AUDIT

Runtime observabilityBUILT INBUILD + PIPE

MANAGED SURFACEREST + SSE or Slack

Create a versioned agent, start subject-scoped sessions, and stream the runtime events your product needs.

Get an API key

03 / OPERATING SURFACE

One dashboard for agents, sessions, and runtime controls.

Build the agent once, deploy it to each customer, and keep every run visible from the same operating surface.

Herm dashboard showing session events, event status, timestamps, and session details.

04 / ACCESS CONTROLS

Permission policies

Set each tool to allow, ask, or deny.

terminal

ask

files

allow

secrets

deny

Role boundaries

Start sessions with the right access level.

readonly

operator

admin

Approval checkpoints

Pause sensitive actions before they run.

tool requestedpaused

human approvalrequired

action runsapproved

Scoped credentials

Attach vault access only where it belongs.

crm vault

slack vault

billing vault

05 / QUESTIONS

Frequently Asked Questions

Everything you need to know about Herm, the API, pricing, and what you can build.

Contact support

What is Herm?

Herm is an API for deploying user-facing AI agents built on the open-source Hermes harness. Create a reusable agent with your system prompt, skills, tools, and connectors, then create a subject-scoped session for each user and chat with it over SSE. We run the containers, the harness, and everything in between.

Do I need to set up any servers or infrastructure?

No. One POST /v1/agents call creates your agent, and Herm runs its sessions in isolated containers with a persistent filesystem and credential management. There's no harness to build, no Docker to wrangle, and no EC2 instance to babysit.

How do I integrate it with my product?

Bring your system prompt, MCP servers, configured Markdown skills, and tools to POST /v1/agents. The API returns an agent ID; create a session with POST /v1/sessions, then wire its SSE events endpoint into your chat UI. Your users get memory, automations, self-learning, and image & video input without you building any of it.

What can my users' agents do?

Anything Hermes can do: browse the web, search, manage files, run your tools, follow your skills, and schedule recurring work. A user can say “every week, look at our top-performing video from last week and make five variations” and the agent handles it in the background.

What happens to my users' files between sessions?

Every agent has a persistent filesystem. Files, memory, skills, and automations survive across sessions, so the agent picks up exactly where it left off — even days later.

Is my users' data private?

Yes. Every user's agent runs in its own isolated container. Secrets are managed through credential references rather than raw tokens, and Herm does not train on your data.

Can I self-host?

Yes. Book a call to discuss private and hybrid deployment options.

06 / DEPLOYMENT

Build your product, not another harness.

Get an API key Read the docs