Skip to content
136.1k
Log inGet started

BlogProduct

Perplexity can now use Graphify

· 3 minute read

Perplexity can now use Graphify. The Graphify skill is live in the Perplexity Computer skills marketplace, open to every Computer user. Type /graphify, paste a link to a public GitHub repository, and Computer maps the whole codebase into a knowledge graph you can explore: every file, every function, and how they connect.

Perplexity is also making tasks that use /graphify free through October 31, 2026, up to 5,000 credits per Computer user.

The Graphify codebase mapped in Perplexity Computer: a force-directed graph of 4,022 nodes, with extract.py selected and a list of code communities beside it.

Perplexity shared the news on X:

Computer (@AskPerplexity) on X: "The Graphify skill by @graphify is now in Computer's skills marketplace, available to all Computer users. Turn a codebase into an interactive map of its files, functions, and dependencies. Tasks that use /graphify are free through October 31, up to 5,000 credits per user."

Computer by Perplexity, on X

Aravind Srinivas (@AravSrinivas) on X: "/graphify now on Computer. We're giving every user 5,000 credits to try this skill that lets you visualize a codebase."

Aravind Srinivas, CEO of Perplexity, on X

Why does this matter for developers?Link to Why does this matter for developers?

Reading a new codebase is slow. You open a file, follow an import, lose the thread, and start again. Graphify gives Computer a map before it answers. It turns a repository into a code knowledge graph, where functions, classes, and files are nodes and calls and imports are the edges between them.

Each edge keeps its direction, what kind of link it is, how confident Graphify is in it, and the file and line it came from. So when Computer explains how a request moves through a system, it can point to where in the code that happens.

Computer runs the open-source Graphify engine and does the reading with its own model. There is nothing to install on your machine.

What happened when we pointed it at our own repo?Link to What happened when we pointed it at our own repo?

We gave Computer one line:

text
map the codebase at https://github.com/Graphify-Labs/graphify

It cloned the repository by itself, picked the 105 files that hold the code and the architecture docs, and came back about five minutes later with the full map: 4,022 nodes, 10,079 directed edges, and 192 communities of closely linked code.

The report read like notes from an engineer who had spent a week in the code:

  • The two most connected functions are _read_text() with 270 links and _make_id() with 234, both in extractors/base.py. Every language extractor builds on them.
  • dispatch_command() in cli.py is the command hub. It calls 127 functions.
  • The pipeline runs detect, extract, build, cluster, analyze, report, export, and Computer matched each step to the function that does it.
  • It even spotted where our architecture doc pointed to the wrong file for the HTML exporter.

Then it suggested a sharp next question about how our deduplication code fits together. Once the map exists, the questions get better.

How do you try Graphify in Perplexity?Link to How do you try Graphify in Perplexity?

Open Plugins, go to Skills, and install Graphify. Then type /graphify in Computer and paste a GitHub link. Runs are free through October 31, per Perplexity's offer. The step-by-step guide walks through every screen.

Can you run Graphify on your own machine too?Link to Can you run Graphify on your own machine too?

Yes. The engine behind the skill is open source on GitHub, and it works in Claude Code, Cursor, Gemini CLI, and the other assistants we support. For a graph your whole team shares, there is Graphify Cloud.

Pick a repository you want to understand, and let Computer draw the map.

Try Graphify now.

Try for free