explore

Navigate and understand codebases using OCXP code graph intelligence and semantic search.

Updated May 12, 2026
One-click install
npx skills add https://github.com/contexthub-md/marketplace --skill explore-contexthub-md
Or copy as Structured Prompt for Agent
Please help me install this Agent Skill.
Skill: explore
Source: https://github.com/contexthub-md/marketplace/tree/main/contexthub/skills/explore
Command: npx skills add https://github.com/contexthub-md/marketplace --skill explore-contexthub-md

SYSTEM DOCUMENTATION & REQUIREMENTS

What problem does it solve? Understanding an unfamiliar codebase is slow and error-prone when done by manually reading files. This Skill gives you structured navigation over an indexed code graph so you can quickly find where something is implemented, trace call chains, and understand architecture. ## Core Features & Use Cases - Multi-mode code search: Route tasks to waypoints, run semantic concept search, graph queries over entity relationships, exact string grep, and filename lookup. - Drill-down inspection: Fetch full entity details, read source files, find references and callers, and follow zoom links for progressive exploration. - Architecture analysis: View service topology, detect module communities, inspect dependency graphs, and read infrastructure diagrams from Terraform. - Use Case: You join a new project and need to know how authentication works. Run the explore workflow to semantically search for auth-related entities, trace their references, and get a summary of key files and relationships. ## Quick Start Ask the assistant to explore how authentication works in this codebase using the ContextHub explore skill.

Frequently Asked Questions about explore

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

FAQPage Schema
How do I find where a feature is implemented in a large codebase?

Use semantic code search over the indexed code graph to locate relevant entities by concept, then drill into entity details and references. The explore workflow combines search, graph queries, and file reads to pinpoint implementation files.

How do I trace a call chain or request flow through code?

Start with a task-based route or concept search to find entry points, then use the references tool to see who calls or imports each entity. Zoom links let you progressively follow the flow from API gateway to database.

What is the difference between semantic search and grep for code exploration?

Semantic search finds code by meaning, such as locating the database connection pool without knowing exact names. Grep performs literal text matching, which is better for exact strings like environment variable names such as AUTH_SECRET.

Why do I get OCXP_UNAVAILABLE or NO_PROJECT errors?

These errors mean no active project with indexed repositories exists. Run the connect skill first to set up a workspace and project, then ensure the repository is indexed before exploring.

Can I analyze architecture and dependencies without reading every file?

Yes. The topology, communities, and dependency graph tools return service dependency maps, detected subsystems, and module boundaries directly from the code graph, plus infrastructure diagrams generated from Terraform.