> ## Documentation Index
> Fetch the complete documentation index at: https://hub.hcompany.ai/llms.txt
> Use this file to discover all available pages before exploring further.

> ## Agent Instructions
> H Platform has four products: the Agents API (managed computer-use agents, base URL https://agp.eu.hcompany.ai/api/v2 or https://agp.hcompany.ai/api/v2 for the US), the Models API (OpenAI-compatible Holo vision-language models at https://api.hcompany.ai/v1), HoloDesktop CLI (Holo on the user's own desktop), and HoloTab (a free no-code Chrome extension that runs Holo in the user's browser, with recordable routines and schedules).
> Authenticate with a bearer API key from the HAI_API_KEY environment variable. SDKs: `pip install hai-agents` (Python, `from hai_agents import Client`) and `npm install hai-agents` (TypeScript, `import { HaiAgentsClient } from "hai-agents"`). CLI: `hai`.
> Agents do work in a browser or on a desktop; describe the task as an imperative instruction. To run a task quickly, prefer the pre-built agent `h/web-surfer-flash`. Read results from the session's `latest_answer` after it reaches a terminal status.
> Sessions are the unit of work; wait for a terminal status (completed, failed, timed_out, interrupted) before reading the answer. Use webhooks or the `changes` long-poll endpoint to follow progress.

# Configuration

> Environment variables, local files, logs, and the runtime cache used by HoloDesktop CLI.

Most users need only `holo login`, or `HAI_AGENT_RUNTIME_BASE_URL` and `HAI_AGENT_RUNTIME_MODEL` to [run the model yourself](/holo-desktop-cli/getting-started/quickstart#run-the-model-yourself). The rest is for agent hosts, debugging, and custom runtimes.

## Environment variables

| Variable | Purpose |
| - | - |
| `HAI_API_KEY` | API key for H's hosted models. |
| `HAI_AGENT_RUNTIME_BASE_URL` | Your OpenAI-compatible model server, such as `http://localhost:8000/v1`. Same as `holo run --base-url`. |
| `HAI_AGENT_RUNTIME_MODEL` | Model ID, such as `holo4-35b-a3b`. Same as `holo run --model`. Required with `HAI_AGENT_RUNTIME_BASE_URL`. |
| `HAI_AGENT_RUNTIME_PORT` | Loopback port of the runtime. Defaults to `18795` for hosted inference; each self-hosted server and model gets its own port. |
| `HAI_AGENT_RUNTIME_API_TOKEN` | Bearer token for the runtime. Generated when unset. |
| `HAI_AGENT_LOCAL_BINARY_PATH` | Run this `hai-agent-runtime` executable. |
| `HAI_AGENT_LOCAL_CACHE_DIR` | Runtime cache. Defaults to `~/.hai/agent-runtime`. |
| `HAI_AGENT_RUNTIME_DOWNLOAD_URL`, `HAI_AGENT_RUNTIME_DOWNLOAD_SHA256` | Download a trusted runtime build. Set both or neither. |

`HAI_API_KEY` is read from, in order: the process environment, `~/.holo/.env`, `.env` in the current directory, then `~/.config/hai/.env`. `XDG_CONFIG_HOME`, when set, replaces `~/.config`.

Apps launched from the Dock or Start menu do not inherit shell exports, so set variables for agent hosts in the host's MCP `env` block. See [Agent hosts](/holo-desktop-cli/integrations/agent-hosts#install-into-a-host).

## Files

`~/.config/hai/` and `~/.hai/agent-runtime/` are shared with the `hai-agents` SDK and the `hai` CLI. Treat all of these as sensitive.

| Path | Purpose |
| - | - |
| `~/.config/hai/.env` | API key saved by `holo login`. |
| `~/.config/hai/stop` | Stop request written by `holo stop` and the double-`Esc` kill switch. A task obeys only a stop filed after it started. |
| `~/.config/hai/desktop.lock` | Held by the agent driving the desktop, so one runs at a time. |
| `~/.holo/agents.md`, `memories.md`, `rules.md`, `holo-memories.md` | Instructions loaded into every run. See [Customize](/holo-desktop-cli/how-to/customize). |
| `~/.holo/skills/` | Skills, bundled and your own. |
| `~/.holo/settings.json` | Which bundled skills were seeded. |
| `~/.holo/.env` | Older key location. Still read, and cleared by `holo logout`. |
| `~/.holo/logs/holo-guard.log` | Kill-switch errors on macOS. |
| `~/.hai/agent-runtime/logs/hai-agent-runtime-<port>.log` | Runtime startup and error log. |
| `~/.hai/agent-runtime/state/agent-token-<port>` | Owner-only token so local clients can share a runtime. Removed when the runtime stops. |
| `~/.hai/agent-runtime/state/agent-inference-<port>` | What that runtime infers against, so a client never attaches to one serving another model. |
| `~/Library/LaunchAgents/ai.hcompany.holo.guard.plist` | macOS: starts the kill switch at login. |
| `holo-guard.cmd` in your Startup folder | Windows: starts the kill switch at login. |
| `~/.config/autostart/holo-guard.desktop` | Linux: starts the kill switch at graphical login. |

## Runtime

The runtime is resolved in this order:

1. `HAI_AGENT_LOCAL_BINARY_PATH`;
2. `hai-agent-runtime` on `PATH`;
3. the managed install under `~/.hai/agent-runtime/bin/<version>/`;
4. a sha256-verified download of the pinned version.

`holo doctor` shows which one it will use. On platforms without a published runtime, put your own `hai-agent-runtime` on `PATH`.


This documentation is built and hosted on [Mintlify](https://mintlify.com), a developer documentation platform.