Investigation: evaluate Graphify for codebase knowledge-graph workflows

[xmanjack]

Context

We've added Graphify as a git submodule in our research repo and want to evaluate it as a knowledge-graph tool that maps code, docs, PDFs, images, and video into a queryable graph.

Goals of this investigation

• Core workflow — Run /graphify . and graphify export callflow-html on representative projects; assess the quality of graph.html, GRAPH_REPORT.md, and graph.json outputs.
• Multi-modal ingestion — Evaluate how well it ingests non-code assets (docs, PDFs, images, video) and how those map into the graph.
• Tooling integration & query — Test integration across our AI coding assistants and assess querying the graph vs. grepping files (the "memory layer" workflow).

Questions to answer

• How accurate and complete is the generated graph for our stack?
• What's the runtime/cost on large repos, and is re-indexing incremental?
• What data leaves the machine, and what are the privacy/security implications?

Deliverable

A short write-up summarizing findings and a go/no-go recommendation.