Source: https://ailoot.ai/docs/integrations/crush/
Canonical: https://ailoot.ai/docs/en/integrations/crush/
Language: en

[Terminal agents](https://ailoot.ai/docs/en/integrations/#agents) 

# Crush

[Markdown](https://ailoot.ai/docs/en/integrations/crush/index.md)

Terminal agent · OpenAI Chat, Anthropic Messages

| Format | Where to set it up |
| --- | --- |
| OpenAI Chat, Anthropic Messages | crushrc (Bash configuration): ./.crushrc, ./crushrc, or ~/.config/crush/crushrc; the legacy crush.json format is deprecated but still works |

## [](https://ailoot.ai/docs/integrations/crush/#setup)Setup

Copy the configuration and replace `sk-...` with your key from the Ailoot Dashboard. Prefer storing the key in the environment variable `AILOOT_API_KEY`, if the application supports it.

Before starting, request GET /v1/models with your Ailoot key. Choose an exact model ID whose methods include POST /v1/chat/completions, POST /v1/messages, then use that ID in the configuration. For agents, also check tool support and the features required by your client version.

`~/.config/crush/crushrc`

```bash
export AILOOT_API_KEY="sk-..."
provider add ailoot --type openai-compat \
  --base-url "https://api.ailoot.ai/v1" \
  --api-key "$AILOOT_API_KEY"

model add ailoot/deepseek-v4-pro \
  --name "DeepSeek V4 Pro (Ailoot)" \
  --context-window 1000000 \
  --default-max-tokens 32000

# Anthropic format for Claude: URL without /v1
provider add ailoot-claude --type anthropic \
  --base-url "https://api.ailoot.ai" \
  --api-key "$AILOOT_API_KEY"

model add ailoot-claude/claude-sonnet-5-5 \
  --name "Claude Sonnet 5.5 (Ailoot)" \
  --context-window 1000000 \
  --default-max-tokens 64000 \
  --can-reason true --supports-images false
```

This configuration reads the key from the environment variable `AILOOT_API_KEY`. Set it before starting the application:

```bash
export AILOOT_API_KEY="sk-..."   # add to ~/.zshrc or ~/.bashrc
```

```powershell
setx AILOOT_API_KEY "sk-..."   # then restart the terminal
```

Example model: `deepseek-v4-pro`. Choose any other model from the [catalog](https://ailoot.ai/models/en/) — check that it supports the format you need.

## [](https://ailoot.ai/docs/integrations/crush/#notes)Good to know

The openai-compat type uses the OpenAI Chat format and a URL with /v1. The anthropic type uses the Messages format and a URL without /v1: Crush passes it to anthropic-sdk-go, which appends /v1/messages. The example in the Crush README with /v1 would produce /v1/v1/messages and a 404 error. The key is sent in the x-api-key header, which Ailoot accepts. If the model list is empty, the openai-compat provider fetches it from /v1/models. crushrc runs as code: do not launch Crush in someone else's directory containing this file. In this example, Claude has supports-images = false: the 2026-09-30 catalog snapshot lists only text for the selected Messages format. If your version of Crush supports this flag, enable it only for a model whose current catalog card lists image input in the selected API format; the client flag alone does not add image support to a model.

## [](https://ailoot.ai/docs/integrations/crush/#check)Test

- For the OpenAI format, the URL is `https://api.ailoot.ai/v1` (or the full path if the application asks for a full URL).
- For the Anthropic format, the URL is `https://api.ailoot.ai` without `/v1`.
- Ask a short question. If you get an error, check the code and Request ID, then consult [Errors and limits](https://ailoot.ai/docs/en/errors/).
- 404 immediately after setup almost always means the URL is wrong: an extra or missing `/v1`.
- 402 — insufficient balance for the reserve. Agents reserve more than chat: [why](https://ailoot.ai/docs/en/billing/reserve/).

Application documentation: [https://github.com/charmbracelet/crush#custom-providers](https://github.com/charmbracelet/crush#custom-providers)

Still have questions? Send your Request ID and a description — we'll help you troubleshoot.[Contact support](mailto:support@ailoot.ai)

## Ailoot · The company and its project portfolio





Ailoot combines established operating experience, a large client base, stable service and ongoing KPI control.





The company and its project portfolio: millions of customers; monthly sales in the billions.





Ailoot brings that operating experience to API access for AI features in customer products, agents, applications and automations.