Project Onboarding
Scenario: A new developer joins the team and needs to understand a project quickly.
The Problem
Onboarding typically involves reading scattered documentation, asking teammates for context, and slowly piecing together how the codebase works. This takes days or weeks, and much of the tribal knowledge never makes it into docs.
The Workflow
With Graph Memory connected to your AI assistant, onboarding becomes a conversation.
1. Explore the Documentation
Start by understanding what documentation exists:
docs_list_files()
This returns all indexed markdown files with their titles and section counts. Get an overview of the project's documentation landscape.
Search for specific topics:
docs_search({ query: "getting started" })
docs_search({ query: "architecture overview" })
2. Understand the Code Structure
List code graph source files (TypeScript/JavaScript) to see how the codebase is organized:
code_list_files()
For a full project structure overview including all file types, use files_list() instead.
Explore symbols in key files:
code_get_file_symbols({ fileId: "src/index.ts" })
code_get_file_symbols({ fileId: "src/api/routes.ts" })
Search for specific concepts in the code:
code_search({ query: "authentication middleware" })
code_search({ query: "database connection" })
3. Read Existing Knowledge
Check for team notes and decisions:
notes_list()
notes_search({ query: "architecture decisions" })
These notes often contain the "why" behind design choices — exactly what onboarding docs miss.
4. Review Current Tasks
See what the team is working on:
tasks_list({ status: "in_progress" })
tasks_list({ status: "todo" })
This gives context on active development areas and priorities.
5. Find Procedures
Check for existing skills and procedures:
skills_list()
skills_recall({ context: "development setup" })
skills_recall({ context: "deployment process" })
Key Tools
| Tool | Purpose in onboarding |
|---|---|
docs_list_files | See all available documentation |
docs_search | Find relevant doc sections by topic |
code_list_files | List code graph source files (TS/JS) |
files_list | Browse full project file tree |
code_get_file_symbols | Explore what a source file exports |
code_search | Find code by concept, not just name |
notes_list | Read team knowledge and decisions |
notes_search | Find notes on specific topics |
tasks_list | See current work items |
skills_recall | Find documented procedures |
Tip: Use the Prompt Builder
The Web UI's Prompt Builder has an Onboarding scenario that generates a system prompt optimized for exploration and discovery. Copy it into your AI assistant configuration to get the best onboarding experience.