Integrated Tools Reference
This page documents all 47 integrated tools in the Hermes tool registry, grouped by toolkit. Availability varies by platform, credentials, and activated toolkits.
Quick count: 10 browser tools, 4 file tools, 10 RL tools, 4 Home Assistant tools, 2 terminal tools, 2 web tools, and 15 standalone tools across other toolkits.
MCP Tools
In addition to built-in tools, Hermes can dynamically load tools from MCP servers. MCP tools appear with the server name prefix (e.g., GitHub_create_issue for MCP server GitHub). See MCP Integration for configuration.
Toolkit browser
| Tool | Description | Environment Requirements |
|---|---|---|
browser_back | Navigate back to the previous page in the browser history. Requires browser_navigate to have been called first. | — |
browser_click | Click on an element identified by its reference ID from a snapshot (e.g., '@e5'). Reference IDs are displayed in square brackets at the beginning of snapshot output. Requires browser_navigate and browser_snapshot to have been called first. | — |
browser_console | Retrieve browser console output and JavaScript errors from the current page. Returns console.log/warn/error/info messages and uncaught JS exceptions. Use this to detect silent JavaScript errors, failed API calls, and application warnings. Requires… | — |
browser_get_images | Get a list of all images on the current page along with their URLs and alt text. Useful for finding images to analyze with the vision tool. Requires browser_navigate to have been called first. | — |
browser_navigate | Navigate to a URL in the browser. Initializes a session and loads the page. Must be called before other browser tools. For simple information retrieval, prefer web_search or web_extract (faster, cheaper). Use browser tools when you need… | — |
browser_press | Press a keyboard key. Useful for submitting forms (Enter), navigation (Tab), or keyboard shortcuts. Requires browser_navigate to have been called first. | — |
browser_scroll | Scroll the page in a direction. Use this to reveal more content that may be below or above the current view. Requires browser_navigate to have been called first. | — |
browser_snapshot | Get a text-based snapshot of the current page's accessibility tree. Returns interactive elements with reference IDs (like @e1, @e2) for browser_click and browser_type. full=false (default): compact view with interactive elements. full=true: fea… | — |
browser_type | Type text into an input field identified by its reference ID. Clears the field first, then types the new text. Requires browser_navigate and browser_snapshot to have been called first. | — |
browser_vision | Capture a screenshot of the current page and analyze it with AI vision. Use this when you need to visually understand content on a page - especially useful for CAPTCHAs, visual verification challenges, complex layouts, or when text is broken… | — |
Toolkit clarify
| Tool | Description | Environment Requirements |
|---|---|---|
clarify | Ask the user a question when you need clarification, feedback, or a decision before proceeding. Supports two modes: 1. Multiple choice — provide up to 4 options. The user selects one or enters their own answer via a 5th 'Other' option. 2.… | — |
Toolkit code_execution
| Tool | Description | Environment Requirements |
|---|---|---|
execute_code | Run a Python script that can programmatically call Hermes tools. Use this when you need more than 3 tool calls with processing logic between them, need to filter/reduce large tool outputs before they enter your context, need conditional branching (… | — |
Toolkit cronjob
| Tool | Description | Environment Requirements |
|---|---|---|
cronjob | Unified scheduled task manager. Use action="create", "list", "update", "pause", "resume", "run" or "remove" to manage jobs. Supports skill-enabled jobs with one or more attached skills and skills=[] on update will remove attached skills. Cron runs occur in new sessions without the current chat context. | — |
Toolkit delegation
| Tool | Description | Environment Requirements |
|---|---|---|
delegate_task | Spawn one or more sub-agents to perform tasks in isolated contexts. Each sub-agent has its own chat, final session, and toolkit. Only the final summary is returned -- intermediate tool results never appear in your context window. TWO… | — |
Toolkit file
| Tool | Description | Environment Requirements |
|---|---|---|
patch | Targeted find-and-replace edits in a file. Use this instead of sed/awk in terminal. Uses fuzzy matching (9 strategies) so minor whitespace/indentation differences won't break it. Returns a unified diff. Automatically runs syntax check after edit… | — |
read_file | Read a text file with line numbers and pagination. Use this instead of cat/head/tail in terminal. Output format: 'LINE_NUM|CONTENT'. Suggests similar filenames if not found. Use offset and limit for large files. NOTE: Cannot read image o… | — |
search_files | Search file contents or find files by name. Use this instead of grep/rg/find/ls in terminal. Powered by Ripgrep, faster than shell equivalent. Content search (target='content'): Regex search inside files. Output modes: full line match… | — |
write_file | Write content to a file, completely replacing existing content. Use this instead of echo/cat heredoc in terminal. Automatically creates parent directories. OVERWRITES entire file - use 'patch' for targeted edits. | — |
Toolkit homeassistant
| Tool | Description | Environment Requirements |
|---|---|---|
ha_call_service | Call a Home Assistant service to control a device. Use ha_list_services to discover available services and their parameters for each domain. | — |
ha_get_state | Get the detailed state of a single Home Assistant entity, including all attributes (brightness, color, temperature setpoint, sensor readings, etc.). | — |
ha_list_entities | List Home Assistant entities. Optionally filter by domain (light, switch, climate, sensor, binary_sensor, cover, fan, etc.) or by area name (living room, kitchen, bedroom, etc.). | — |
ha_list_services | List available Home Assistant services (actions) for controlling devices. Shows what actions can be performed on each device type and what parameters they accept. Use this to discover how to control devices found via ha_list_entities. | — |
ghi chú
Honcho tools (Honcho_conclude, Honcho_context, Honcho_profile, Honcho_search) are no longer built-in. They are available via the Honcho memory provider plugin at plugins/memory/Honcho/. See Plugins for installation and usage.
Toolkit image_gen
| Tool | Description | Environment Requirements |
|---|---|---|
image_generate | Generate high-quality images from text prompts using the FLUX 2 Pro model with automatic 2x upscaling. Creates detailed, artistic images that are automatically upscaled for high-resolution results. Returns a single upscaled image URL. Display it using… | FAL_KEY |
Toolkit memory
| Tool | Description | Environment Requirements |
|---|---|---|
memory | Save important information to persistent memory that exists across sessions. Your memory appears in the system prompt at the start of a session -- it's how you remember things about the user and your environment between conversations. WHEN TO SA… | — |
Toolkit messaging
| Tool | Description | Environment Requirements |
|---|---|---|
send_message | Send a message to a connected messaging platform or list available targets. IMPORTANT: When the user requests sending to a specific channel or person (not just a bare platform name), call send_message(action='list') FIRST to see available tar… | — |
Toolkit moa
| Tool | Description | Environment Requirements |
|---|---|---|
mixture_of_agents | Route a difficult problem through multiple frontier LLMs collaboratively. Makes 5 API calls (4 reference models + 1 aggregator) with maximum reasoning effort - use sparingly for truly hard problems. Best for: complex math, advanced algorithms… | OpenRouter_API_KEY |
Toolkit rl
| Tool | Description | Environment Requirements |
|---|---|---|
rl_check_status | Get status and metrics for a training run. RATE LIMIT: minimum 30 minutes between checks for the same run. Returns WandB metrics: step, status, reward_avg, loss, accuracy_percent. | TINKER_API_KEY, WANDB_API_KEY |
rl_edit_config | Update a configuration field. First use rl_get_current_config() to see all available fields for the chosen environment. Each environment has different configuration options. Infrastructure settings (token, URL, lora_rank, learning_ra… | TINKER_API_KEY, WANDB_API_KEY |
rl_get_current_config | Get the current environment configuration. Returns only modifiable fields: group_size, max_token_length, Total_steps, step_per_eval, use_wandb, Wanb_name, max_num_workers. | TINKER_API_KEY, WANDB_API_KEY |
rl_get_results | Get final results and metrics for a completed training run. Returns final metrics and path to trained weights. | TINKER_API_KEY, WANDB_API_KEY |
rl_list_environments | List all available RL environments. Returns environment name, path, and description. TIP: Read the file_path with file tools to understand how each environment works (verifier, data loading, rewards). | TINKER_API_KEY, WANDB_API_KEY |
rl_list_runs | List all training runs (active and completed) along with their status. | TINKER_API_KEY, WANDB_API_KEY |
rl_select_environment | Select an RL environment for training. Loads the environment's default configuration. After selecting, use rl_get_current_config() to view settings and rl_edit_config() to modify them. | TINKER_API_KEY, WANDB_API_KEY |
rl_start_training | Start a new RL training process with the current environment and configuration. Most training parameters (lora_rank, learning_rate, etc.) are fixed. Use rl_edit_config() to set group_size, batch_size, Wanb_project before starting. WARNING: Training… | TINKER_API_KEY, WANDB_API_KEY |
rl_stop_training | Stop a running training job. Use if metrics look bad, training is stalled, or you want to try different settings. | TINKER_API_KEY, WANDB_API_KEY |
rl_test_inference | Quick inference test for any environment. Performs a few inference + scoring steps via OpenRouter. Default: 3 steps x 16 completions = 48 deployments per model, testing 3 models = 144 total. Tests environment loading, quick crafting, in… | TINKER_API_KEY, WANDB_API_KEY |
Toolkit session_search
| Tool | Description | Environment Requirements |
|---|---|---|
session_search | Search your long-term memory for past conversations. This is your recall function -- every previous session is searchable and this tool summarizes what happened. ACTIVELY USE THIS when: - The user says 'we did this before', 'remember when', 'last time… | — |
Toolkit skills
| Tool | Description | Environment Requirements |
|---|---|---|
skill_manage | Manage skills (create, update, delete). Skills are your procedural memory -- reusable approaches for recurring task types. New skills go to ~/.Hermes/skills/; existing skills can be modified wherever they live. Actions: create (full SKILL.m… | — |
skill_view | Skills enable loading information about specific tasks and workflows as well as scripts and templates. Load the entire skill content or access its associated files (references, templates, scripts). The first call returns SKILL.md content plus… | — |
skills_list | List available skills (name + description). Use Skill_view(name) to load full content. | — |
Toolkit terminal
| Tool | Description | Environment Requirements |
|---|---|---|
process | Manage background processes started by terminal(background=true). Actions: 'list' (show all), 'poll' (check status + new output), 'log' (full output with pagination), 'wait' (block until complete or timeout), 'kill' (terminate), 'write' (sen… | — |
terminal | Execute shell commands on a Linux environment. The filesystem persists between calls. Set background=true for long-running servers. Set notify_on_complete=true (with background=true) to receive automatic notification when the process ends - no polling needed. DO NOT use cat/head/tail - use read_file. DO NOT use grep/rg/find - use search_files. | — |
Toolkit todo
| Tool | Description | Environment Requirements |
|---|---|---|
todo | Manage your task list for the current session. Use for complex tasks with more than 3 steps or when the user provides multiple tasks. Call with no parameters to read the current list. Write: - Provide 'todos' array to create/update items - merge=… | — |
Toolkit vision
| Tool | Description | Environment Requirements |
|---|---|---|
vision_analyze | Analyze an image with AI vision. Provides a comprehensive description and answers specific questions about the image content. | — |
Toolkit web
| Tool | Description | Environment Requirements |
|---|---|---|
web_search | Search the web for information on any topic. Returns up to 5 relevant results with title, URL, and description. | EXA_API_KEY or PARALLEL_API_KEY or FIRECRAWL_API_KEY or TAVILY_API_KEY |
web_extract | Extract content from a webpage URL. Returns page content in markdown format. Also works with PDF URLs - pass the PDF link directly and it will convert to markdown text. Pages under 5000 characters return full markdown; larger pages are LLM-summarized. | EXA_API_KEY or PARALLEL_API_KEY or FIRECRAWL_API_KEY or TAVILY_API_KEY |
Toolkit tts
| Tool | Description | Environment Requirements |
|---|---|---|
text_to_speech | Convert text to speech audio. Returns a MEDIA: path that the platform delivers as a voice message. On Telegram, it plays as a voice bubble; on Discord/WhatsApp as an audio attachment. In CLI mode, saves to ~/voice-memos/. Voice and provider… | — |