Published by Gigabrain Labs · Updated
What you need before you start
- Install a file-capable agent and check its setup requirements on the compatibility page.
- Choose a dedicated project folder and back up existing material.
- Decide which sources the agent may read and which external services it may use.
- After purchase, verify the download checksum and open START-HERE in your licensed archive.
Run a small first task
- Begin with fictional notes or material you are comfortable sharing with your agent.
- Ask one question with a known answer.
- Request the supporting source.
- Add a correction to the notes.
- Repeat the question in a fresh session.
- Ask a second question the notes cannot answer.
- Check whether the agent explains that the answer is missing.
Repeat these checks with the tasks you plan to use Gigabrain for.
Check the instructions for your agent
Follow the instructions shipped with your release. Agent versions and adapter behavior change. The compatibility page distinguishes tested automation from manual invocation. Do not enable scheduled or provider-backed work until you understand its permissions.
Local and connected operations
| Operation | Where it runs / network use | What to verify |
|---|---|---|
| Source files and notes | Files in your chosen local project or vault. | Remote sync, backup and shared folders have their own destinations. |
| Agent reasoning | Your chosen agent; a hosted agent sends selected context to its provider. | A local vault does not make a hosted agent offline. |
| Retrieval | Search and citation checks run locally after model setup. | Hosted agents can still send the context you share to their provider. |
| Graph authoring | Optional cloud graph authoring can send selected source excerpts to your chosen provider. | Check which provider receives the material and what you are approving before enabling it. |
| Scheduled operations | An enabled schedule can execute approved operations without another prompt. | Check what runs, where data goes, how often it runs and how to stop it. |
| Updates and remote Git | Downloading updates or sending changes to remote Git storage needs network access. | Choose your remote and inspect changes before applying them. |
Check your installation
Start in a fresh extracted copy of your purchased archive. Python 3.10 or newer is required. Windows setup uses Git Bash. See the compatibility page for the operating systems covered by our automation tests.
sh bootstrap.sh
"$(cat .gigabrain/python)" tools/gigabrain_doctor.py --preflight
"$(cat .gigabrain/python)" tools/gigabrain_doctor.pyBootstrap records the interpreter. Use it for subsequent commands. Follow the delivered README for the one-time local retrieval-model download, ingestion and indexing. Before installing an adapter, review the proposed changes. Back up your existing agent configuration.
Your first evidence check
Use the supplied fictional garden-handbook sample in the delivered README. Ask: “Ask my vault how often the Thorne Street beds are watered, and cite it.” Open the cited file and passage. Repeat in a new session. If retrieval is missing or outdated, run the doctor. Repair the index before accepting an answer. Agent subscriptions, model access and optional Obsidian are separate from the Gigabrain licence.
Start with one useful answer
Open QUICK-START in your download with your chosen AI agent. Choose one document and a useful question that it can answer. The guided setup helps prepare the file, build its index and check the quoted source. Connect more information once that first task works.
Keep your cloud storage
You can keep using Google Drive, Dropbox, OneDrive or iCloud. Tell your agent which application and account you use. CONNECT-YOUR-SOURCES explains connections, exports and files that need downloading. A chat connection alone does not add documents to the local index.
Read your knowledge in Obsidian
We recommend Obsidian for browsing your notes and their links. It is optional. Your AI agent still runs Gigabrain.
Download Obsidian. Follow our opening and graph-view guide.
