[Integrations](https://spiral.ai/integrations) › [Hermes](https://spiral.ai/integrations/hermes) › [GPT Astra](https://spiral.ai/integrations/hermes/gpt-astra)

# How to Use GPT Astra in Hermes

GPT Astra suits complex coding, research, and document work with several requirements to track. Give it a clear deliverable and verification criteria so it can carry the work through to a reviewable result.

|  |  |
| --- | --- |
| Model | GPT Astra |
| Available through | OpenAI |
| Category | LLM |
| Price | Up to $0.26681 per run at the default inputs. $13.00 per 1M input tokens, $1.30 per 1M cached input tokens, $16.25 per 1M cache write tokens, $65.00 per 1M output tokens. |
| Returns | text, json |
| Runs | In the background |
| Setup | About 2 minutes, no API key |

## Prompts to try with GPT Astra in Hermes

- Every Monday, use GPT Astra to turn the supplied project updates into a weekly brief. Cover completed work, changed commitments, blockers, and decisions needed; cite the update behind each point and keep the brief under 700 words.
- Each morning, use GPT Astra to review the supplied research updates against our stated research question. Report new evidence, contradictions, and unanswered questions, distinguishing today's changes from previously reported findings.
- Every Friday, use GPT Astra to check the supplied procurement requests against the written approval policy. Return each request's expected approval path, supporting rule, and missing information; flag ambiguous cases for review.
- Each week, use GPT Astra to draft a six-slide status presentation from the supplied updates. Follow the established outline, preserve reported numbers, and flag inconsistencies before presenting conclusions.
- At the end of each day, use GPT Astra to compare the supplied workflow results with their acceptance criteria. Report unmet requirements and evidence gaps, make only explicitly permitted assumptions, and finish with a review queue.

## How to set up GPT Astra in Hermes

Pick the way that suits you. Each takes about two minutes.

### Use the Spiral skill

1. **Send Hermes this message**: Your agent reads Spiral's skill and sets itself up.

```
set up https://spiral.ai/SKILL.md
```

2. **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

1. **Add the server and sign in**: Hermes opens your browser: sign in to Spiral and click Authorize. No API key. When Hermes asks Enable all tools? \[Y/n/select], answer Y: n cancels without saving the server, and select hides the tools you leave out.

```bash
hermes mcp add spiral \
  --url https://api.spiral.ai/mcp \
  --auth oauth \
  --connect-timeout 300
```

2. **Check it**: In a chat that was already open, run `/reload-mcp`.

```bash
hermes mcp list
hermes mcp test spiral
```

### Use the Spiral CLI

1. **Install and sign in**: Installs the Spiral CLI, signs you in and adds its skill to your coding agents.

```bash
npx -y spiral-cli@latest init
```

### Call the Spiral API

1. **Create an API key**: In the console, under API keys.

2. **Run GPT Astra**: One request; the result comes back by polling or a webhook.

```bash
curl https://api.spiral.ai/v1/runs \
  -H "Authorization: Bearer $SPIRAL_API_KEY" \
  -H "Content-Type: application/json" \
  -H "Idempotency-Key: $(uuidgen)" \
  -d '{
    "provider": "openai",
    "model": "~openai/gpt-astra-latest",
    "input": {
      "messages": [
        {
          "role": "user",
          "content": "Hello!"
        }
      ]
    }
  }'
```

Then paste one of the prompts above: Hermes finds GPT Astra on Spiral, shows the price and runs it.

Setup guide in the docs: https://spiral.ai/docs/quickstarts/hermes

## GPT Astra tools and prices

- [GPT Astra Latest](https://spiral.ai/tools/openai/~openai/gpt-astra-latest) (OpenAI): Up to $0.213447 per run at the default inputs. $10.40 per 1M input tokens, $0.52 per 1M cached input tokens, $13.00 per 1M cache write tokens, $52.00 per 1M output tokens.

## Settings you can ask for

| Setting | Allowed values | Default | Description |
| --- | --- | --- | --- |
| seed |  | — |  |
| tools |  | — |  |
| reasoning |  | — | Reasoning effort; reasoning tokens count toward max_tokens. |
| max_tokens | 16 to 128000 | — | The most output tokens to generate. |
| tool_choice |  | — |  |
| response_format |  | — |  |

## What can you do with GPT Astra in Hermes?

GPT Astra can turn recurring updates into research briefs and checks against written business rules. Define the reporting period, evidence requirements, and decisions that need human review.

### Debug a failure across several files

Have GPT Astra trace a reported failure through the relevant code, propose a focused fix, and verify it against a reproducible test.

### Plan a database migration

Ask GPT Astra to turn schema changes and operational constraints into a migration plan with compatibility checks, rollback conditions, and unresolved decisions.

### Write a research decision brief

Give GPT Astra source material and a concrete decision. Request a brief that connects supporting evidence, disagreements, and information still needed.

### Draft a report in an established format

Supply a report template, source text, and audience. Ask GPT Astra to preserve the section order and style while filling gaps only where the evidence supports them.

### Develop a presentation narrative

Use GPT Astra to turn research notes into a slide-by-slide argument, with a clear takeaway and supporting evidence for each slide.

### Review a business approval process

Ask GPT Astra to check written approval rules against normal cases and exceptions. Request a table of expected outcomes for human review.

### Extract evidence into structured records

Have GPT Astra convert supplied text into JSON records with evidence excerpts and explicit missing values. Validate the records against the schema and your business rules.

## Overview

- [OpenAI](https://spiral.ai/providers/openai)
- [All Hermes guides](https://spiral.ai/integrations/hermes)

## GPT Astra in other agents

- [Amp](https://spiral.ai/integrations/amp/gpt-astra)
- [ChatGPT](https://spiral.ai/integrations/chatgpt/gpt-astra)
- [Claude](https://spiral.ai/integrations/claude-desktop/gpt-astra)
- [Claude Code](https://spiral.ai/integrations/claude-code/gpt-astra)
- [Cline](https://spiral.ai/integrations/cline/gpt-astra)
- [Codex](https://spiral.ai/integrations/codex/gpt-astra)
- [Cursor](https://spiral.ai/integrations/cursor/gpt-astra)
- [Devin](https://spiral.ai/integrations/devin/gpt-astra)
- [dots](https://spiral.ai/integrations/dots/gpt-astra)
- [Factory Droid](https://spiral.ai/integrations/factory-droid/gpt-astra)
- [Gemini](https://spiral.ai/integrations/gemini/gpt-astra)
- [Gemini CLI](https://spiral.ai/integrations/gemini-cli/gpt-astra)
- [Google Antigravity](https://spiral.ai/integrations/antigravity/gpt-astra)
- [goose](https://spiral.ai/integrations/goose/gpt-astra)
- [Grok](https://spiral.ai/integrations/grok/gpt-astra)
- [Grok Build](https://spiral.ai/integrations/grok-build/gpt-astra)
- [Instinct](https://spiral.ai/integrations/instinct/gpt-astra)
- [Manus](https://spiral.ai/integrations/manus/gpt-astra)
- [Muse](https://spiral.ai/integrations/muse/gpt-astra)
- [Muse Code](https://spiral.ai/integrations/muse-code/gpt-astra)
- [n8n](https://spiral.ai/integrations/n8n/gpt-astra)
- [Notion](https://spiral.ai/integrations/notion/gpt-astra)
- [OpenClaw](https://spiral.ai/integrations/openclaw/gpt-astra)
- [opencode](https://spiral.ai/integrations/opencode/gpt-astra)
- [Perplexity](https://spiral.ai/integrations/perplexity/gpt-astra)
- [Raycast](https://spiral.ai/integrations/raycast/gpt-astra)
- [VS Code](https://spiral.ai/integrations/vscode/gpt-astra)
- [Windsurf](https://spiral.ai/integrations/windsurf/gpt-astra)
- [Zapier](https://spiral.ai/integrations/zapier/gpt-astra)

## More Hermes guides

- [GLM-5V-Turbo](https://spiral.ai/integrations/hermes/glm-5v-turbo)
- [GPT Chat Latest](https://spiral.ai/integrations/hermes/gpt-chat-latest)
- [GLM-5.3](https://spiral.ai/integrations/hermes/glm-5-3)
- [GPT Luna Latest](https://spiral.ai/integrations/hermes/gpt-luna-latest)
- [GLM-5.2](https://spiral.ai/integrations/hermes/glm-5-2)
- [GPT Mini Latest](https://spiral.ai/integrations/hermes/gpt-mini-latest)

## Questions about GPT Astra in Hermes

**Can Hermes use GPT Astra?**

Yes. Add Spiral to Hermes as an MCP server, and Hermes can find and run GPT Astra like any built-in tool.

**How much does GPT Astra cost in Hermes?**

Up to $0.26681 per run at the default inputs. $13.00 per 1M input tokens, $1.30 per 1M cached input tokens, $16.25 per 1M cache write tokens, $65.00 per 1M output tokens. You pay from your Spiral balance as you go, with no subscription, and a failed run is not charged. Your Hermes plan is separate.

**Do I need an OpenAI account to use GPT Astra in Hermes?**

No. Spiral runs GPT Astra on its own OpenAI account. You sign in to Spiral once and never paste a key.

**How long does it take to set up GPT Astra in Hermes?**

About two minutes: add the server and sign in, then check it.

**Does Hermes wait while GPT Astra runs?**

GPT Astra runs in the background. Hermes starts the run, checks on it and gives you the text, json when it's ready.

**What is GPT Astra good at?**

GPT Astra suits complex coding, research, and document creation. Use it when the task requires connecting evidence, tracking requirements, and producing a finished deliverable.

**What are Astra's context and output limits?**

OpenAI documents a 1,050,000-token context window and a maximum of 128,000 output tokens for GPT-6 Astra. Internal reasoning also consumes the output budget, so that maximum does not mean 128,000 tokens of visible text.

**How should I prompt GPT Astra?**

State the deliverable, constraints, desired format, and checks that define success. Use direct instructions; you do not need to ask it to reveal its internal reasoning.

**When should I adjust reasoning effort?**

Use lower effort for straightforward follow-ups where speed matters. For difficult work, test higher effort against clear acceptance criteria and leave enough output space for both reasoning and the answer.

**Why does GPT Astra sometimes ask a question instead of finishing?**

Astra may request clarification when missing information could change the result. Tell it which assumptions it may make, which decisions require your input, and what it should finish before stopping.

