Bot Commands

/helpShow available commands

/newStart a new session (fresh session ID + history)

/stopKill all running background processes

/statusShow session, model, token, and context info

/resumeResume a previously-named session

/sessionsBrowse and resume previous sessions

/modelSwitch model (persists by default)

/debugUpload debug report (system info + logs) and get shareable links

/restartGracefully restart the gateway after draining active runs

/updateUpdate Hermes Agent to the latest version

/commandsBrowse all commands and skills (paginated)

/approveApprove a pending dangerous command

/denyDeny a pending dangerous command

/queueQueue a prompt for the next turn (doesn't interrupt)

/steerInject a message after the next tool call without interrupting

/backgroundRun a prompt in the background

/reasoningManage reasoning effort and display

/usageShow token usage and rate limits for the current session

/platformPause, resume, or list a failing gateway platform

/profileShow active profile name and home directory

/whoamiShow your slash command access (admin / user)

/startAcknowledge platform start pings without a reply

/topicEnable or inspect Telegram DM topic sessions

/retryRetry the last message (resend to agent)

/undoBack up N user turns and re-prompt (default 1)

/titleSet a title for the current session

/branchBranch the current session (explore a different path)

/compressCompress conversation context (add 'here [N]' to keep recent N turns; --preview shows what would happen)

/rollbackList or restore filesystem checkpoints

/agentsShow active agents and running tasks

/goalSet a standing goal Hermes works on across turns until achieved

/moaRun one prompt through the default Mixture of Agents preset, then restore your model

/subgoalAdd or manage extra criteria on the active goal

/sethomeSet this chat as the home channel

/codex_runtimeToggle codex app-server runtime for OpenAI/Codex models

/personalitySet a predefined personality

/footerToggle gateway runtime-metadata footer on final replies

/yoloToggle YOLO mode (skip all dangerous command approvals)

/fastToggle fast mode — OpenAI Priority Processing / Anthropic Fast Mode (Normal/Fast)

/voiceToggle voice mode

/memoryReview pending memory writes / toggle the approval gate

/bundlesList skill bundles (aliases /<name> for multiple skills)

/learnLearn a reusable skill from anything you describe (dirs, URLs, this chat, notes)

/suggestionsReview suggested automations (accept/dismiss)

/blueprintSet up an automation from a blueprint template

/curatorBackground skill maintenance (status, run, pin, archive, list-archived)

/kanbanMulti-profile collaboration board (tasks, links, comments)

/reload_mcpReload MCP servers from config

/reload_skillsRe-scan ~/.hermes/skills/ for newly installed or removed skills

/creditsShow Nous credit balance and top up

/insightsShow usage insights and analytics

/versionShow Hermes Agent version

HermesBuddy

Bot Commands