> For the complete documentation index, see [llms.txt](https://notara.gitbook.io/notara-docs/llms.txt). Markdown versions of documentation pages are available by appending `.md` to page URLs; this page is available as [Markdown](https://notara.gitbook.io/notara-docs/byok/anthropic.md).

# Anthropic (Claude)

Anthropic's Claude models are the recommended choice for Notara. Claude excels at following complex instructions, multi-step reasoning, long context understanding, and using tools accurately — all of which matter when an agent is running tasks on your team's behalf.

## Getting Your API Key

1. Go to [console.anthropic.com](https://console.anthropic.com).
2. Sign in or create an Anthropic account.
3. Navigate to **API Keys** in the left sidebar.
4. Click **Create Key**.
5. Give it a descriptive name (e.g., "Notara production").
6. Copy the key — it starts with `sk-ant-` and will only be shown once.

Store the key somewhere secure before navigating away.

## Adding the Key to Notara

1. In the Notara dashboard, go to **Settings → BYOK**.
2. Select **Anthropic** from the provider dropdown.
3. Paste your `sk-ant-...` key.
4. Click **Verify** — you'll see a green checkmark if the key is valid.
5. Click **Save**.

If you see an error, double-check:

* No leading or trailing whitespace
* The key starts with `sk-ant-api03-` (for recent keys) or `sk-ant-`
* Your Anthropic account has billing set up (keys without billing configured will fail test calls)

## Which Model Notara Uses

Notara uses **Claude 3.5 Sonnet** by default for the primary agent. This is the sweet spot between performance and cost — it's fast, handles long contexts well, and uses tools reliably.

For high-complexity workspaces (many tool calls, long conversations, complex reasoning tasks), Notara may escalate to **Claude 3 Opus**. You can configure the preferred model in **Settings → BYOK → Model Preference**.

## Cost Estimation

Anthropic charges per token (input and output separately). Current pricing (verify at [anthropic.com/pricing](https://www.anthropic.com/pricing)):

| Model             | Input (per 1M tokens) | Output (per 1M tokens) |
| ----------------- | --------------------- | ---------------------- |
| Claude 3.5 Sonnet | \~$3.00               | \~$15.00               |
| Claude 3 Opus     | \~$15.00              | \~$75.00               |
| Claude 3 Haiku    | \~$0.25               | \~$1.25                |

With 1 Notara credit = 10,000 tokens, and assuming a mix of input/output:

* **Pro plan (2,500 credits / 25M tokens)** at Sonnet pricing ≈ $40–$80/month in Anthropic costs, depending on the input/output ratio.
* Most teams use far fewer than 25M tokens per month; 5–10M is typical for active teams.

Your actual Anthropic costs appear in the [Anthropic console](https://console.anthropic.com/settings/billing) under usage.

## Rate Limits

Anthropic applies rate limits per API key based on your usage tier. New accounts start at Tier 1 (limited RPM and TPM). As you spend more, Anthropic automatically upgrades your tier.

If Notara agents are hitting rate limit errors, you'll see them reflected in the agent's Slack responses ("I encountered a rate limit, please try again in a moment"). To increase your limits:

1. Go to [console.anthropic.com/settings/limits](https://console.anthropic.com/settings/limits).
2. Check your current tier and usage.
3. Anthropic upgrades tiers automatically based on spend history — sustained usage will eventually move you up.

For teams with large-scale usage, contact Anthropic directly about enterprise rate limit tiers.

## Rotating Your Key

To rotate your Anthropic API key (recommended periodically or if you suspect a leak):

1. Generate a new key in the Anthropic console.
2. In Notara, go to **Settings → BYOK**, paste the new key, and click **Save**.
3. Once verified, the old key is no longer stored. Delete it from the Anthropic console.

The rotation is atomic — the new key takes effect immediately for all subsequent agent calls.

## Context Window

Claude 3.5 Sonnet has a 200,000 token context window. Notara's context budget system ensures this is never exceeded: spine documents are injected up to a budget, and the rest are listed by name for the agent to request on demand. Long Slack conversation histories are truncated from the oldest messages first.


---

# Agent Instructions
This documentation is published with GitBook. GitBook is the documentation platform designed so that both humans and AI agents can read, navigate, and reason over technical content effectively. Learn more at gitbook.com.

## Querying This Documentation
If you need additional information that is not directly available in this page, you can query the documentation dynamically by asking a question.

Perform an HTTP GET request on the current page URL with the `ask` query parameter, and the optional `goal` query parameter:

```
GET https://notara.gitbook.io/notara-docs/byok/anthropic.md?ask=<question>&goal=<endgoal>
```

`ask` is the immediate question: it should be specific, self-contained, and written in natural language.
`goal` is optional and describes the broader end goal you are ultimately trying to accomplish on behalf of the user. GitBook uses it to tailor the answer towards what is most useful for that goal.

The response will contain a direct answer to the question and relevant excerpts and sources from the documentation.

Use this mechanism when the answer is not explicitly present in the current page, you need clarification or additional context, or you want to retrieve related documentation sections.