graphify-sf

Parse local Salesforce SFDX directories and map metadata relationships with provenance tags.

4|Updated May 12, 2026
One-click install
npx skills add https://github.com/raykuonz/graphify-sf --skill graphify-sf
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: graphify-sf
Source: https://github.com/raykuonz/graphify-sf/tree/main/skills/graphify-sf
Command: npx skills add https://github.com/raykuonz/graphify-sf --skill graphify-sf

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve?

Large Salesforce orgs store critical metadata dependencies across thousands of scattered XML files and Setup screens, making it extremely time-consuming to trace relationships like which trigger fires on a specific object, which Flow writes to a field an Apex class reads, or which Profile grants access to unused metadata. Grep and string search only find text matches, not actual dependency relationships, and manual Setup navigation is impractical for large projects.

Core Features & Use Cases

  • Fully Offline Metadata Parsing: Extracts all SFDX metadata entities (Apex classes, triggers, Flows, Custom Objects, LWC/Aura components, Profiles, Permission Sets, etc.) directly from local source files with no org connection or SFDX CLI required.
  • Provenance-Tagged Relationship Mapping: Maps real dependencies between metadata, with every edge tagged as EXTRACTED (from explicit XML) or INFERRED (from source patterns) so you always know if a relationship is a confirmed fact or a best guess.
  • Use Case: Use this skill for impact analysis before making metadata changes, tracing dependency chains for debugging, exploring org architecture for onboarding, or generating ground truth data for AI coding agents working with Salesforce projects.

Quick Start

Use the graphify-sf skill to build a queryable knowledge graph of your current Salesforce SFDX project directory.

Frequently Asked Questions about graphify-sf

High-intent search queries and answers about installing and using this skill.

FAQPage Schema
How do I trace Salesforce metadata dependencies without connecting to an org?

You can trace Salesforce metadata dependencies offline by parsing local SFDX source directories to extract entities and map their relationships. This approach requires no org connectivity or SFDX CLI, directly reading XML files and source code to build a queryable knowledge graph.

What is the best way to do impact analysis for Salesforce Apex classes and Flows?

Impact analysis for Apex classes and Flows is best handled by generating a knowledge graph that maps explicit and inferred relationships. This tags every dependency edge with provenance, showing exactly which metadata is affected by a proposed change.

Can I use grep to find which Flow writes to a field an Apex class reads?

Grep cannot effectively find which Flow writes to a field an Apex class reads because it only matches text, not actual dependency relationships. Parsing SFDX source files maps real dependencies across metadata, identifying true structural connections.

Does graphify-sf work with local SFDX project files for offline metadata analysis?

Yes, graphify-sf works with local SFDX project files for offline metadata analysis by extracting entities like Custom Objects, LWC components, and Profiles directly from source. It generates interactive visualizations and GraphRAG-ready JSON exports without org access.

How do I know if a mapped Salesforce dependency is a confirmed fact or an inferred guess?

To know if a mapped Salesforce dependency is confirmed or inferred, check the provenance tags on each relationship edge. The knowledge graph explicitly tags every connection as EXTRACTED from XML or INFERRED from source patterns.

What formats does the SFDX metadata knowledge graph export generate?

The SFDX metadata knowledge graph export generates interactive visualizations, GraphRAG-ready JSON exports, and plain-language reports. These outputs map all metadata entities and their relationships with full provenance tagging.