Keys no longer have Standard or Prompt modes. A server prompt is configured independently for Claude, Kimi, GLM, Gemini, Muse, Composer, Grok, and Codex. The model field selects the model; the key and USD balance remain shared.
If any condition is missing, the request proceeds without a server prompt. Entitlement and the option control application only. Messages, Chat Completions and Responses use the selected model’s full rates; the separate native Composer/Grok path uses a preliminary debit cap.
| Surface | What counts as input | Charge |
|---|---|---|
| Messages, Chat Completions and Responses without a server prompt | Client input | Full input or cache-read rate |
| Messages, Chat Completions and Responses with a server prompt | Client input and server-prompt tokens | Full input or cache-read rate |
| Ordinary API: usage or available count_tokens | Full input_tokens, including an applied server prompt | The prompt text is not needed |
/cursor-native/v1, Composer/Grok | Preliminary input and server-prompt estimate | Reserve and maximum debit |
- Check the selected surface: Messages, Chat Completions or Responses. A shared address does not mean identical parameters.
- Claude has specific thinking, media and citation behavior when a server prompt is applied; the detailed matrix is on the Messages page.
- The server prompt does not work on every model; those it works on are marked UNLOCK in Guard Code. The other models work as usual for a key with the prompt: without the prompt and without an error.
An ordinary model refusal is returned in the response. Guard does not stop replies with a local refusal or repeated-text detector. Request limits and error handling still apply.
A structured Claude cybersecurity refusal returns the cyber_refusal code:
Your Guard account and the Usage API show request history, actual server-prompt application, full token usage and cost.
Claude returns tool_use, the client executes the function, and the client sends the result as tool_result. Guard carries that exchange but does not execute the tool on its own server.
tool_result content | Handling |
|---|---|
String or text | Passed to the model as text |
image | Base64 is supported; a URL is not fetched for a regular key, so download a remote image and send it as base64 |
document | Text sources and base64 PDFs are passed to the model as documents; shared server-prompt media and size limits apply |
search_result | Passed as a client search result and cannot be mixed with other result types in one result |
tool_reference | Passed as a reference to a client-side tool |
is_error: true | Passed to the model as a tool-execution error |
{
"role": "user",
"content": [{
"type": "tool_result",
"tool_use_id": "toolu_…",
"content": [
{"type": "text", "text": "Check completed"},
{
"type": "image",
"source": {
"type": "base64",
"media_type": "image/png",
"data": "…"
}
}
]
}]
}Choose the existing guide for your client. Installation steps are not duplicated here.
anthropic and openaiNode / TypeScriptExamples for both SDKsn8nConnect automationsMigrate from OpenAIMove an OpenAI SDK client to GuardOn Messages, Chat Completions and Responses, token cost is calculated at the requested model’s full published rate, including cache writes and reads. An applied server prompt is part of billable input. Fixed server-tool fees are added after the token portion. Composer/Grok on /cursor-native/v1 uses the preliminary debit cap described above.
- Server-prompt entitlement and the selected option affect application but do not reduce the rate.
- Active promotions, package bonuses and credit already granted remain separate from token cost.
- The new rules do not recalculate historical financial records.
| Setting | What happens |
|---|---|
| Key, address and balance | Stay the same |
| Family | Claude, Kimi, GLM, Gemini, Muse, Composer, Grok, and Codex have independent options and entitlements |
| Request already started | Retains its initial prompt application and price |
| Next request | Uses the current entitlement and option |
| Entitlement expiry | The chosen option is retained; application stops and resumes after renewal |
- Open @guardapi → My API keys and choose the key.
- Open key management and the intended family: Claude, Kimi, GLM, Gemini, Muse, Composer, Grok, or Codex.
- Check the server-prompt entitlement, its expiry and the selected option.
- After changing the setting, send a new request and check actual application in its history.
- The key card shows the option and entitlement for the intended family.
- Your Guard account shows actual application, full token usage and the cost of each request.
- In Usage API, use
familyandendpoint; request records includeserver_prompt_appliedanddiscount_percent. Thediscount_percentfield remains for historical records and separate promotions; no automatic discount is represented as0. The removedmodefilter returns410 product_modes_removed.
- Server-prompt settings affect subsequent requests. An already started stream retains its original conditions.
- Start a new agent session after changing its server-prompt binding.
- Choose the exact
modelin your client; a server-prompt entitlement does not choose the family for you.
- Reopen the key card if you are viewing an old bot message.
- Check the key, family, entitlement expiry and enabled option.
- Check a new request: a setting change does not affect a stream already in progress.
- If the entitlement is active and the option is enabled but no prompt was applied, ask support whether a suitable prompt is available. Do not send your full API key.
Codex has its own entitlement and per-key option; neither changes the four existing families. A prompt is applied only when that entitlement is active, the option is enabled, and a suitable Codex binding exists.
The Grok server prompt applies in Chat Completions, Messages and Responses when the entitlement is active, the Grok option is enabled and a prompt is available. Choose the model with the same model field, for example cursor-grok-4.7-high. Responses and history keep the requested model, and billing uses its rates.
Images, client functions and multi-turn dialogue work the same as without a server prompt. The separate /cursor-native/v1 contour is unchanged.
The GLM server prompt is sold in @Guardpromptbot as server access and is not issued as a file. It applies when the key has an active entitlement (a purchase, PRIVATE or a manual grant from Guard), the key's toggle is on and the server-side text is available. Connecting the entitlement to a key turns the toggle on; you can turn it off in your Guard account. With an applied prompt, GLM requests run at temperature 1.0 and the client's value is not used.
The Gemini server prompt is configured like that of the other families: the key needs an active entitlement for the Gemini family and the application option turned on. Without them, the request goes without a prompt.
The Muse server prompt is configured like that of the other families: the key needs an active entitlement for the Muse family and the application option turned on. Without them, the request goes without a prompt.