KabuForge
Menu
GitHub日本語

Ask an AI Agent to Help Install KabuForge: Scope and Checks

Define the selected version, installation paths and approval scope, then verify what actually ran.

Check official entry points and choose a version

Start with the official repository and Get started guide. Decide whether to use a published release or the development main branch, then ask the assistant to read the README for that selection. A moving branch should not be treated as a fixed published distribution.

An assistant with local terminal and file access may help with installation. In a chat-only environment, you perform the suggested actions yourself. Asking AI does not itself complete installation or connection; check the available capabilities and approved scope.

Specify paths and proposed changes

Identify the OS, Python, existing environments, installation location and research directory. Ask for a dedicated virtual environment and a proposal listing dependencies, sources, changed files and required permissions. Use the selected version's requirements and preserve existing environments.

Approve the concrete targets and operations before execution. Client access settings and platform approval controls do not mean KabuForge automatically grants permission. Check that the requested scope and the operations actually permitted agree before changes begin.

Begin with a short scoped request

Replace the paths and selected version in the request below with explicit choices for your environment. Installation and MCP connection are separate tasks. A small first request covering inspection, a change proposal, approved installation and diagnostics makes failures easier to review.

Installation approval does not automatically authorize private-data access, external transmission, additional downloads, paid services or research execution. If needed, explain and approve each target and purpose separately. Never paste secret values into the request.

Official repository: https://github.com/Siyuan-chat/KabuForge
Official getting-started guide: https://kabuforge.com/en/get-started/
Help me install KabuForge.
Read the official repository and the selected version's README; inspect my OS, Python, installation path and dedicated environment.
Show changed files, downloaded dependencies and required permissions. Execute only after I approve that scope.
Check doctor and the official synthetic demo in that environment; report completed and unperformed checks.
Wait for separate approval before MCP configuration changes, private-data access, external transmission or research execution.

Reconcile diagnostics and the synthetic demo

Run kabuforge doctor from the selected environment and verify the version, executable location and available capabilities. Then follow that version's official guide for its synthetic offline demo, checking outputs and exit status. Synthetic fixtures do not establish market investment performance.

The final report should identify checks actually executed, paths, output identities, errors and omitted checks. Do not use a source-run screenshot as proof that a different distribution was installed successfully. Investigate errors before expanding dependencies.

Connect MCP separately and verify it

After installation checks, use the existing guide for Codex, Claude, Cursor or stdio. Confirm absolute executable and workspace paths, review proposed client settings and approve them before applying changes.

Verify tool listing and authorized read-only inspection. A listed tool is not a successful research run. Record restarts and manual steps still required, and do not report an unconnected configuration as a completed connection.

Related reading and official sources

Read Point-in-time data · Lookahead bias · What is backtesting? to place this topic alongside other research checks. Confirm documented operations and limits in the official installation documentation · published releases. Use documentation for your selected build; if public main differs from your environment, defer unverified claims.

This article teaches research procedures. Numerical examples are fictional teaching illustrations, not actual investment performance. It does not recommend securities or trades or guarantee future returns. Historical public courses do not establish complete historical information availability. Disclaimer

Continue this topic

Explore the research data learning path

Applies to: main source candidate (unpublished) · Public source docs/en_US/README.md

Reviewed · KabuForge