How to Use Grok 4.5 in Instinct
Grok 4.5 suits repository debugging, technical analysis, and tasks that require several tool calls. Its scope extends beyond coding to data science and business writing.
Prompts to try with Grok 4.5 in Instinct
Copy an example into Instinct, or ask in your own words.
On each daily run, use Grok 4.5 with medium reasoning effort to compare the supplied service metrics with yesterday's values. Report significant changes and calculations supporting each finding.
Every Monday, use Grok 4.5 to turn the supplied weekly results into a five-slide briefing outline. Include evidence for each claim and list unanswered questions.
On each run, use Grok 4.5 to extract dates, organizations, and commitments from newly supplied text. Return consistent records with evidence excerpts and mark absent values as unknown.
Each morning, use Grok 4.5 with high reasoning effort to analyze the supplied failure logs. Rank possible causes and propose a test that could confirm each one.
Every Friday, use Grok 4.5 to compare the supplied technical updates with the previous report. Summarize changed assumptions, new evidence, and decisions needing review.
How to set up Grok 4.5 in Instinct
Pick the way that suits you. Each takes about two minutes.
Use the Spiral skill
- Send Instinct this message
Your agent reads Spiral's skill and sets itself up.
set up https://spiral.ai/SKILL.md
- Sign in or paste a key
Approve Spiral in the browser, or give the agent a key from the console.
Use Spiral's MCP server
- Create an API key
In the console, open API keys and create a key for Instinct. Copy it.
- Send Instinct this message
Instinct reads Spiral's skill and asks for your API key.
set up https://spiral.ai/SKILL.md
- Give Instinct the key
When Instinct asks for the key, add it where Instinct keeps secrets, never in the chat itself.
Use the Spiral CLI
- Install and sign in
Installs the Spiral CLI, signs you in and adds its skill to your coding agents.
npx -y spiral-cli@latest init
Call the Spiral API
- Create an API key
In the console, under API keys.
- Run Grok 4.5
One request; the result comes back by polling or a webhook.
curl https://api.spiral.ai/v1/runs \ -H "Authorization: Bearer $SPIRAL_API_KEY" \ -H "Content-Type: application/json" \ -H "Idempotency-Key: $(uuidgen)" \ -d '{ "provider": "openrouter", "model": "x-ai/grok-4.5", "input": { "messages": [ { "role": "system" } ] } }'
Then paste one of the prompts above: Instinct finds Grok 4.5 on Spiral, shows the price and runs it.
Grok 4.5 providers and prices
Spiral runs Grok 4.5 through OpenRouter. Open a card for today's price, with every fee included.
Up to $0.03204 per run at the default inputs. $2.60 per 1M input tokens, $7.80 per 1M output tokens.
Settings you can ask for
Ask Instinct for any of these in your prompt.
| Setting | Allowed values | Default | Description |
|---|---|---|---|
| Seed | a whole number | ||
| Top p | a number up to 1 | ||
| Max tokens | 1 to 450000 | The most output tokens to generate. | |
| Temperature | 0 to 2 | ||
| Reasoning effort | low, medium, or high | high | How much the model reasons. |
Every API field of SpaceXAI: Grok 4.5 is on its tool page: open its card above.
What can you do with Grok 4.5 in Instinct?
Turn each scheduled run's evidence into a consistent technical or business report with Grok 4.5. Define what to compare and what evidence each finding must include.
Fix repository bugs
Ask Grok 4.5 to trace a failing behavior through the code, propose a patch, and verify it against a regression test.
Debug Rust and C/C++
Give Grok 4.5 a compiler error and the relevant code. Request a correction that preserves the intended ownership rules or memory behavior.
Build an application prototype
Turn a defined workflow into a working application with explicit acceptance criteria. Have Grok 4.5 check the main user path before reporting completion.
Investigate data anomalies
Have Grok 4.5 compare reported figures, identify inconsistencies, and propose calculations that distinguish competing explanations.
Extract document records
Turn supplied text into structured records with Grok 4.5. Specify required values, evidence excerpts, and how to mark missing information.
Draft business presentations
Give Grok 4.5 the approved findings and ask for slide titles, supporting evidence, and speaker notes for a defined audience.
Compare technical proposals
Ask Grok 4.5 to compare approaches against explicit engineering constraints, expose assumptions, and identify the tests needed to make a decision.
Produce recurring analytical reports
Have Grok 4.5 turn each run's supplied observations into a consistent report of changes, evidence, unresolved questions, and next actions.
More on Grok 4.5 and Instinct
Overview
Grok 4.5 in other agents
More Instinct guides
Questions about Grok 4.5 in Instinct
Can Instinct use Grok 4.5?
Yes. Add Spiral to Instinct as an MCP server, and Instinct can find and run Grok 4.5 like any built-in tool.
How much does Grok 4.5 cost in Instinct?
Up to $0.03204 per run at the default inputs. $2.60 per 1M input tokens, $7.80 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Instinct plan is separate.
Do I need an OpenRouter account to use Grok 4.5 in Instinct?
You need no OpenRouter account. Instinct uses one Spiral API key for every tool, and Spiral runs Grok 4.5 on its own OpenRouter account.
How long does it take to set up Grok 4.5 in Instinct?
About two minutes: create an API key, then send Instinct this message, then give Instinct the key.
Does Instinct wait while Grok 4.5 runs?
Grok 4.5 runs in the background. Instinct starts the run, checks on it and gives you the text, json when it's ready.
What is Grok 4.5 best suited for?
Use it for software engineering and analytical work that requires several steps, including debugging, data analysis, and comparing technical approaches. Give it a concrete deliverable and a way to check the result.
How much context can Grok 4.5 handle?
The maker lists a 500,000-token context window. For a large task, identify the relevant sections and the exact question you want answered.
How should I choose the reasoning effort?
Start with low for straightforward work, medium for analysis, and high for difficult debugging or mathematical problems. Grok 4.5 defaults to high and always uses reasoning.
Can Grok 4.5 produce structured records?
Yes. Document extraction and report generation are useful applications: define the record you need and tell it how to represent missing information.
How do I prompt Grok 4.5 effectively?
State the task, constraints, and desired output directly. For example, ask for a patch that preserves a public interface, includes a regression test, and explains the cause in three sentences.
Start using Grok 4.5 in Instinct
One Spiral account for every tool your agent needs. No subscription; you pay per run.
Paste this into Instinct:
set up https://spiral.ai/SKILL.md


