G
Graphify
graphify.homes
Sign in
Gemini CLI context

Gemini CLI codebase context with a queryable graph

Gemini CLI can use repository guidance, tools, and project files while working, but long sessions still need a durable way to navigate structure. Graphify adds graph.html, graph.json, and GRAPH_REPORT.md so Gemini can query dependencies and architecture instead of rereading the repository on every task.

Published Updated

Graphify outputs
graph.html
graph.json
GRAPH_REPORT.md

Graphify stores inspectable artifacts so codebase knowledge can be reviewed by humans and reused by AI coding assistants.

A Gemini CLI terminal connected to a constellation of project graph nodes

Guidance

GEMINI.md

Use repository instructions for commands, conventions, protected paths, and tool policy.

Structure

Graphify

Use graph artifacts for dependencies, calls, subsystem communities, and documentation links.

Interface

CLI or MCP

Query the graph through the Graphify skill/CLI or expose it as MCP tools.

What is Gemini CLI codebase context?

Gemini CLI codebase context is the information Gemini uses from the repository, configured instructions, command output, and connected tools. The quality depends on finding the right structure and respecting project rules before editing.

  • Repository files provide implementation detail.
  • GEMINI.md provides durable project guidance.
  • Graphify provides persistent structural relationships.

What belongs in GEMINI.md?

GEMINI.md should contain actionable instructions: build and test commands, coding conventions, protected paths, review requirements, and when to query the graph. Keep it concise so it remains stable as the repository grows.

  • Document the repository's source of truth.
  • State when graph queries are required.
  • Require source verification before code changes.

How does Graphify improve Gemini CLI navigation?

Graphify gives Gemini a project map before it reads full files. Gemini can follow relationships, inspect god nodes and communities, and use report sections to decide which source files deserve detailed reading.

  • Trace paths between services, files, and symbols.
  • Use communities to explain subsystem boundaries.
  • Connect code to docs, PDFs, screenshots, and rationale.
  • Refresh the graph after major changes.

How do you install Graphify for Gemini CLI?

Use graphify gemini install, then build a graph in the repository. The installer writes the Gemini integration and guidance needed to invoke Graphify from the CLI session.

  • Install Python 3.10+ and the graphifyy package.
  • Run graphify gemini install.
  • Build the first graph with /graphify or the documented CLI workflow.

When should Gemini use MCP instead of reading files?

Use MCP when the question requires a repeatable graph operation such as search, path traversal, explanation, or community context. Read files when exact implementation details or recent edits matter.

  • MCP returns focused graph results.
  • Source files remain authoritative for behavior.
  • Tests verify the proposed change.

How should teams evaluate the setup?

Measure whether Gemini finds relevant files faster, follows fewer irrelevant paths, and produces review findings with stronger source evidence. Do not use token count alone as a quality metric.

  • Compare answer accuracy on representative architecture questions.
  • Track repeated discovery and file reads.
  • Verify that graph freshness is part of the workflow.
Working commands

How do the setup commands fit together?

Install the Gemini CLI integration

Shell
graphify gemini install
FAQ

Frequently asked questions

Does Graphify replace GEMINI.md?

No. GEMINI.md holds project policy and instructions; Graphify holds queryable project structure.

Can Gemini CLI use both Graphify and MCP?

Yes. The skill/installer provides workflow guidance, and MCP exposes structured graph operations.

Does Gemini need to upload the repository?

Graphify's code extraction is designed to run locally. Review model configuration and provider policy when adding docs or images that require semantic extraction.

How often should the graph refresh?

Refresh after structural changes, major refactors, dependency updates, and before architecture reviews.

Key takeaways

What should you remember?

  • Use GEMINI.md for working rules and Graphify for persistent structure.
  • The gemini installer connects the workflow; MCP provides structured tool calls.
  • Verify graph-derived findings against source and tests.
Sources

Where can you verify the details?

Gemini CLI GitHub repositoryCurrent CLI capabilities, configuration, and extension surfaces.Graphify GitHub repositoryGemini installer, graph artifacts, MCP, and query workflows.Graphify architecture documentationHow extracted code structure and semantic context are connected.

Related Graphify guides

Continue through the Graphify topic cluster with pages that explain assistant context, architecture review, documentation, and graph-based codebase understanding.

GitHub Copilot codebase context beyond file search

Give GitHub Copilot durable codebase context with repository instructions, local graph artifacts, and MCP tools for dependencies and architecture questions.

MCP codebase knowledge graph for AI coding assistants

Use an MCP codebase knowledge graph to give Codex, Claude, and other agents structured queries for dependencies, architecture, and project context.

Codex codebase context without rereading the repository

Give Codex persistent codebase context with graph artifacts, concise AGENTS.md rules, and MCP tools so repository questions use structure, not repeated scans.

Turn this concept into a real project graph

Start from the use cases page or upload a project ZIP to build a graph that your team can inspect and reuse.

See use casesBuild a graph