Question 1

What is Octopus?

Accepted Answer

Octopus is an AI-powered code review tool. It connects to your GitHub or Bitbucket repositories, indexes your codebase, and automatically reviews pull requests — posting findings as inline comments with severity levels.

Question 2

How does Octopus review my code?

Accepted Answer

When a pull request is opened, Octopus fetches the diff, retrieves relevant context from your indexed codebase using vector search, and sends it to an LLM (Claude or OpenAI) for analysis. The results are posted as PR comments with severity indicators: 🔴 Critical, 🟠 Major, 🟡 Minor, 🔵 Suggestion, 💡 Tip.

Question 3

Which programming languages does Octopus support?

Accepted Answer

Octopus is language-agnostic. It reviews any text-based code file — TypeScript, Python, Go, Rust, Java, C#, Ruby, PHP, Swift, Kotlin, and more. Since it uses LLMs for analysis, it understands the semantics and patterns of virtually any language.

Question 4

Does Octopus replace human code reviewers?

Accepted Answer

No. Octopus is designed to augment your team's review process, not replace it. It catches bugs, security issues, and style inconsistencies so your human reviewers can focus on architecture, design decisions, and business logic.

Question 5

How is Octopus different from linters or static analysis tools?

Accepted Answer

Linters check syntax and formatting rules. Static analysis tools look for known code patterns. Octopus goes further — it understands your codebase context, reads your documentation and coding standards, and provides semantic-level feedback similar to what an experienced developer would give.

Question 6

Is my code safe?

Accepted Answer

Yes. Your code is processed in-memory and never stored permanently on our servers. Embeddings are stored in your Qdrant instance for search, but the original source code is not retained. If you self-host, everything stays on your own infrastructure.

Question 7

Can I self-host Octopus?

Accepted Answer

Absolutely. Octopus is open source (MIT license) and fully self-hostable. You can deploy it with Docker on your own infrastructure — your code never leaves your servers. See the Self-Hosting documentation for setup instructions.

Question 8

Which AI models process my code?

Accepted Answer

Octopus supports Anthropic Claude and OpenAI models. You can configure which model your organization uses, or bring your own API keys (BYO keys) so requests go directly to the provider without any intermediary.

Question 9

Does Octopus train AI models on my code?

Accepted Answer

No. Your code is never used for training. When using the cloud service, code is sent to Anthropic or OpenAI via their API, which does not use API inputs for model training. When self-hosting, you control the entire pipeline.

Question 10

Which Git platforms are supported?

Accepted Answer

Octopus integrates with GitHub (including GitHub Enterprise) and Bitbucket. It installs as a GitHub App or connects via Bitbucket OAuth, and listens for pull request webhooks.

Question 11

Can I connect Octopus to Slack?

Accepted Answer

Yes. The Slack integration lets you ask questions about your codebase directly in Slack channels. Octopus uses your indexed code and knowledge base to provide context-aware answers.

Question 12

Does Octopus integrate with Linear?

Accepted Answer

Yes. You can connect Octopus to Linear to automatically create issues from review findings. This makes it easy to track and prioritize the issues Octopus discovers.

Question 13

Is there a CLI?

Accepted Answer

Yes. The Octopus CLI lets you trigger reviews, index repositories, check usage, and chat with your codebase — all from the terminal. Install it with npm install -g @octp/cli.

Question 14

How does pricing work?

Accepted Answer

Octopus uses a credit-based system. Each review, indexing operation, and chat message consumes credits based on token usage. You can purchase credits or bring your own API keys to use your existing provider billing directly.

Question 15

Is there a free tier?

Accepted Answer

Yes. Every organization gets free credits to get started. This is enough to try Octopus on a few repositories and see the value before committing to a paid plan.

Question 16

What are BYO (Bring Your Own) keys?

Accepted Answer

BYO keys let you use your own Anthropic or OpenAI API keys. This way, AI usage is billed directly to your provider account, and you only pay Octopus for the platform — not the AI tokens.

Question 17

Can I set spend limits?

Accepted Answer

Yes. Each organization can configure monthly spend limits. Octopus checks the limit before every expensive operation and stops processing if you're over budget — no surprise bills.

Question 18

How does codebase indexing work?

Accepted Answer

Octopus fetches your repository contents via the Git provider API, splits the code into overlapping chunks (1500 characters with 200 character overlap), generates vector embeddings, and stores them in Qdrant. The original source code is processed in-memory and not retained — only the embeddings are persisted. This enables fast, semantic code search during reviews.

Question 19

What is the Knowledge Base?

Accepted Answer

The Knowledge Base lets you upload documents (coding standards, architecture docs, style guides) that Octopus references during reviews. This helps it give feedback that's consistent with your team's specific conventions and requirements.

Question 20

How long does a review take?

Accepted Answer

Most reviews complete in 30–90 seconds depending on the size of the diff and the complexity of the changes. Large PRs with hundreds of files may take a few minutes.

Question 21

Does Octopus support real-time updates?

Accepted Answer

Yes. The dashboard uses WebSocket connections to provide real-time updates when reviews complete, repositories are indexed, or new findings are discovered. You'll see results as they happen.

Frequently Asked Questions

General

Security & Privacy

Integrations

Pricing & Billing

Technical