Hermes command cheat sheet

The commands you actually use, on one page. Click any command to copy it.

Reflects the official Hermes Agent docs and GitHub README as read on 2026-10-20. The latest stable release at that time was v0.22.2 (Oct 15, 2026), with nightly canary builds on top. Commands and flags change quickly between releases, so check with hermes <command> --help.

Install & update

Install on Linux, macOS or WSL2
Install on native Windows (PowerShell)
Update a source install to the latest stable release
Show how far behind your channel you are, without updating
Show the installed version
Preview uninstall; plain hermes uninstall keeps your data, --full removes everything

Everyday chat

Start an interactive session (classic CLI)
Start the newer TUI
Seed a session with a prompt
Scripted one-shot that prints only the final reply
Auto-approve actions (skips approval prompts). Skips every safety prompt: use only in a sandbox.
Run a shell command from the prompt without a model turn
Insert a newline in the input
Clear the draft, or interrupt the agent on an empty prompt

Slash commands in chat

Show available commands
Start a fresh session (alias /reset)
Show or change the model
Remove the last user/assistant exchange
Show a breakdown of context-window usage
Show token usage and cost
Apply a personality overlay (e.g. concise, pirate; none clears)
Run a prompt in a background session
Ask a side question without interrupting
List or restore filesystem checkpoints
Exit the CLI (alias /exit)

Models & config

Interactive provider and model picker
Full setup wizard
One-shot Nous Portal login plus Tool Gateway
Set a value; dotted keys go to config.yaml, UPPER_SNAKE keys to .env
Print one value (secrets masked unless --raw)
Find missing options after an update
Run agent commands inside Docker instead of locally
Add a key to a provider's credential pool

Sessions

Resume the most recent session (short: -c)
Resume a specific session (short: -r)
List sessions with title, preview and last-active time
Rename a session
Export one session as markdown (bulk exports need a filter)
Preview deleting old ended sessions (90-day default cutoff)

Memory

Configure an external memory provider
Timeline of learned skills and memories (aliases: learning, memory-graph)
List staged memory writes awaiting approval
Apply staged memory writes (or approve <id>)
Run the memory/skill review now

Skills

Browse hub skills
Preview a skill before installing
Install a skill (with security scan)
List installed skills
Reinstall changed hub skills
Preload a skill for the session
Load an installed skill and add your instruction
Turn a file, URL, workflow or notes into a skill
Rescan skills and refresh their slash commands

Tools & MCP

Configure enabled tools per platform (interactive)
Enable specific toolsets for a run
List available tools in the CLI
Interactive MCP server picker
Install a catalog MCP server (e.g. deepwiki)
Run Hermes as an MCP server
Reload MCP servers after config changes
Run Hermes as an ACP server for editor integration

Gateway & messaging

Configure messaging platforms interactively
Run the gateway in the foreground
Install as a user service (systemd on Linux, launchd on macOS)
Start the background service (also stop, restart)
Check gateway service status
View pending and approved users
Approve a DM pairing code
In a chat: mark it as the home channel for deliveries
In a chat: approve a pending dangerous command (/deny rejects)

Scheduling (cron)

Create a recurring job
Show scheduled jobs
Change a job's schedule
Pause a job (resume to restart)
Trigger a job on the next tick
Delete a job
Create a one-shot job from chat
Global emergency stop for cron, kanban dispatch and gateway turns (hermes resume to undo)

Profiles

Create an isolated profile and its <name> command alias
Copy config, .env, SOUL.md, skills and memory from the current profile
List profiles and their status
Set the sticky default profile ("default" switches back)
Target a profile for one command
Pack a profile into a .tar.gz archive (credentials stripped; -o sets the path)

Diagnostics

Diagnose config and dependency issues (--fix attempts repairs)
Print every status section (--deep runs slower checks)
Follow the agent log (names: agent, errors, gateway, mcp, ...)
Print a copy-pasteable setup summary for support
Token, cost and activity analytics
Back up the Hermes home directory to a zip

Key files and folders

Path	What it holds
`~/.hermes/`	Hermes home and default profile; override with HERMES_HOME
`~/.hermes/config.yaml`	Main non-secret settings: model, terminal backend, memory, skills, compression, display, MCP servers
`~/.hermes/.env`	API keys and other secrets
`~/.hermes/auth.json`	OAuth credentials (e.g. Nous Portal)
`~/.hermes/SOUL.md`	Primary agent identity, loaded first in the system prompt
`~/.hermes/memories/MEMORY.md`	Agent's own notes; ~2,200 character limit
`~/.hermes/memories/USER.md`	Your user profile; ~1,375 character limit
`~/.hermes/skills/<category>/<skill>/SKILL.md`	Installed and agent-created skills
`~/.hermes/state.db`	SQLite session database (messages, FTS5 search index) for CLI, TUI and gateway
`~/.hermes/sessions/`	Gateway session files and `/save` snapshots (sessions/saved/)
`~/.hermes/cron/jobs.json`	Scheduled job definitions; run output in `~/.hermes/cron/output/`
`~/.hermes/logs/`	Logs such as errors.log and gateway.log, with secrets redacted
`~/.hermes/profiles/<name>/`	A named profile's own config, .env, memories, skills, sessions and cron
`~/.hermes/hermes-agent/`	Source checkout from the POSIX install script
`~/.local/bin/hermes`	CLI entry point (POSIX script install); profile aliases also live in `~/.local/bin/`
`%LOCALAPPDATA%\hermes\`	Windows source-install data directory (code in hermes-agent\, CLI in bin\)
`<repo>/.hermes/skills/`	Project-local skills (must be trusted first)