How do I install the Graphify skill?

Run npx skillmds@latest add manastalukdar/graphify in your terminal (requires Node.js), paste this page's agent-chat prompt into Claude, Cursor, or any MCP-connected agent, or download the SKILL.md file and copy it into your agent's skills directory.

Is Graphify safe to use?

This skill has not completed SkillMD's automated safety review yet. Capability flags: makes network calls, reads secrets. SkillMD never runs a skill's scripts for you; review the SKILL.md before installing.

Which AI agents work with Graphify?

This skill is tagged as working with Claude Code, Claude.ai, OpenAI Codex. SKILL.md is an open format, so most agents that read a skills directory can load it too.

Is Graphify free to use?

Yes. Installing skills from SkillMD is free, and the skill stays under its author's original license.

Who published Graphify?

manastalukdar (@manastalukdar) published this skill. Their other Agent Skills are listed on their SkillMD profile.

Graphify by manastalukdar — Agent Skill

Name: Graphify
Rating: 4.2 (5 reviews)
Author: manastalukdar

File contents

Graphify — Codebase Knowledge Graph

Turn the project into a queryable knowledge graph. Delegates to the Graphify CLI (graphifyy on PyPI).

Usage

/graphify              # graph the current directory
/graphify <path>       # graph a specific subdirectory
/graphify --callflow   # also export a Mermaid call-flow HTML page

Behavior

Step 1 — Verify Graphify is installed

graphify --version 2>/dev/null || echo "NOT_INSTALLED"

If not installed, stop and print:

Graphify is not installed. Install it with:

    uv tool install graphifyy        # recommended (auto-adds to PATH)
    pipx install graphifyy           # alternative

Then re-run /graphify.

Step 2 — Run the graph

Use the path from $ARGUMENTS if provided, otherwise .:

TARGET="${ARGUMENTS:-.}"
graphify "$TARGET"

Step 3 — Report outputs

After the command succeeds, report the three output files:

graphify-out/
├── graph.html       Open in any browser — click nodes, filter, search
├── GRAPH_REPORT.md  Key concepts, surprising connections, suggested questions
└── graph.json       Full graph — query without re-reading files

If --callflow was passed, also run:

graphify export callflow-html

and report the additional output.

Step 4 — Suggest next steps

Offer one follow-up based on what the graph found:

If GRAPH_REPORT.md mentions orphaned modules → suggest /understand --deep
If it highlights a complex service → suggest /architecture-diagram
Otherwise → suggest opening graph.html and querying a key concept

Edge Cases

No source files found: Graphify will warn; relay the message and suggest running from the repo root.
Large monorepo: Pass a subdirectory (e.g., /graphify src/) to limit scope.
Stale graph: Delete graphify-out/ and re-run to force a full rebuild.
PowerShell: Remind Windows users to run graphify . not /graphify ..

Token Optimization

Expected range: 200–400 tokens (the skill itself); Graphify runs out-of-process so no additional LLM tokens are consumed.

Patterns used: Bash for system queries (version check, CLI invocation); no file reads; no caching needed — Graphify manages its own output directory.

Early exit: Returns immediately with install instructions if the CLI is missing, saving all downstream tokens.

---name: graphifydescription: Build a queryable knowledge graph from the current codebase using the Graphify CLI — code, SQL schemas, docs, and more in one interactive graph---# Graphify — Codebase Knowledge GraphTurn the project into a queryable knowledge graph. Delegates to the [Graphify](https://github.com/safishamsi/graphify) CLI (`graphifyy` on PyPI).## Usage```/graphify # graph the current directory/graphify <path> # graph a specific subdirectory/graphify --callflow # also export a Mermaid call-flow HTML page```## Behavior### Step 1 — Verify Graphify is installed```bashgraphify --version 2>/dev/null || echo "NOT_INSTALLED"```If not installed, stop and print:```Graphify is not installed. Install it with: uv tool install graphifyy # recommended (auto-adds to PATH) pipx install graphifyy # alternativeThen re-run /graphify.```### Step 2 — Run the graphUse the path from `$ARGUMENTS` if provided, otherwise `.`:```bashTARGET="${ARGUMENTS:-.}"graphify "$TARGET"```### Step 3 — Report outputsAfter the command succeeds, report the three output files:```graphify-out/├── graph.html Open in any browser — click nodes, filter, search├── GRAPH_REPORT.md Key concepts, surprising connections, suggested questions└── graph.json Full graph — query without re-reading files```If `--callflow` was passed, also run:```bashgraphify export callflow-html```and report the additional output.### Step 4 — Suggest next stepsOffer one follow-up based on what the graph found:- If `GRAPH_REPORT.md` mentions orphaned modules → suggest `/understand --deep`- If it highlights a complex service → suggest `/architecture-diagram`- Otherwise → suggest opening `graph.html` and querying a key concept## Edge Cases- **No source files found**: Graphify will warn; relay the message and suggest running from the repo root.- **Large monorepo**: Pass a subdirectory (e.g., `/graphify src/`) to limit scope.- **Stale graph**: Delete `graphify-out/` and re-run to force a full rebuild.- **PowerShell**: Remind Windows users to run `graphify .` not `/graphify .`.## Token Optimization**Expected range**: 200–400 tokens (the skill itself); Graphify runs out-of-process so no additional LLM tokens are consumed.**Patterns used**: Bash for system queries (version check, CLI invocation); no file reads; no caching needed — Graphify manages its own output directory.**Early exit**: Returns immediately with install instructions if the CLI is missing, saving all downstream tokens.

Frequently asked questions

Graphify