Getting Started: Engineers
Get up and running with the Forward Impact CLI tools. This guide covers browsing your career framework, generating AI agent teams, interpreting engineering artifacts with Guide, and setting up your personal knowledge base.
Install Pathway
npm install @forwardimpact/pathway
Initialize framework data
If your organization hasn't provided a framework data bundle, bootstrap starter data to explore with:
npx fit-pathway init
This creates ./data/pathway/ with a complete starter
framework. If your organization distributes a framework bundle,
follow their installation instructions instead — typically a
one-line curl | bash install script that places data at
~/.fit/data/pathway/.
Data directory resolution
The CLI resolves the data directory by walking upward from the
current working directory looking for a
data/pathway/ folder. To override, use the
--data flag:
npx fit-pathway discipline --list --data=./my-data/pathway
Browse your job definition
Use the Pathway CLI to explore the engineering framework your organization has defined.
npx fit-pathway discipline --list # See available disciplines
npx fit-pathway level --list # See available levels
npx fit-pathway track --list # See available tracks
Generate a complete job definition by combining a discipline, level, and optional track:
npx fit-pathway job software_engineering L3 --track=platform
This produces a full view of the skills, behaviours, and expectations for that role.
Generate agent teams
Create AI agent definitions matched to your role's skill profile:
npx fit-pathway agent software_engineering --track=platform --output=./agents
This generates a set of .agent.md files and a
skills/ directory. Each agent file defines a persona
with specific capabilities. The skills directory contains
SKILL.md files that agents use as operational context —
the same skill definitions humans reference, formatted for AI
consumption.
Copy the output into your project's .claude/ or
equivalent agent configuration directory.
Set up Guide
Guide is a conversational AI agent that understands your organization's engineering framework. It helps you onboard, find growth areas, and interpret engineering artifacts against your skill markers.
Install
npm install @forwardimpact/guide
npx fit-codegen --all
The fit-codegen step generates gRPC service clients
that Guide needs. Without it, imports fail with a missing module
error.
Prerequisites
Guide connects to the Forward Impact knowledge platform services. You need a running service stack and a service secret:
export SERVICE_SECRET=<your-secret>
If your organization hosts the platform, they will provide the
service secret and endpoint configuration. If
SERVICE_SECRET is not set, the CLI prints onboarding
instructions.
Usage
npx fit-guide # Start interactive conversation
echo "What skills should I focus on for L3?" | npx fit-guide # Pipe a question
Guide reasons about your organization's specific skill definitions, behaviour expectations, and markers — not generic career advice.
Set up Basecamp
Basecamp is your personal operations center. It syncs email and calendar, builds a knowledge graph, drafts responses, and prepares meeting briefings.
Note: Basecamp requires Bun 1.2+ as its runtime.
Install
bun install @forwardimpact/basecamp
Initialize a knowledge base
bunx fit-basecamp --init ~/Documents/Personal
Check status
bunx fit-basecamp --status
Run the scheduler
bunx fit-basecamp --daemon
Basecamp runs as a macOS status menu app with scheduled AI tasks handling background work. The CLI scheduler works on any platform.
Next steps
- Agent teams -- configure and customize generated agents
- Finding your bearing -- Guide usage and configuration
- Knowledge systems -- deep dive into Basecamp features
- Career paths -- understand progression and skill development
- CLI reference -- full command documentation for all tools