Using SageOx with Claude Code
SageOx is designed to work with AI coding agents. Claude Code is the primary integration — every session starts with your team's full context, and every session's decisions flow back into your knowledge base.
How it works
When you start Claude Code in a SageOx-connected repo, the agent automatically runs ox agent prime. This injects your Team Context — conventions, architectural decisions, domain terminology, and insights from past discussions and recordings.
Your agent doesn't start from zero. It starts from where your team left off.
That's it. No configuration needed beyond ox init.
What your agent receives
| Context type | Source | Example |
|---|---|---|
| Team norms | AGENTS.md in Team Context | "We use snake_case for all API fields" |
| Architectural decisions | Recorded discussions | "We chose Postgres JSONB for metadata because the schema isn't stable yet" |
| Domain terminology | Team Context definitions | "A 'parcel' is a geographic land unit, not a shipping package" |
| Recent decisions | Transcribed walkthroughs | "The upload flow was redesigned last week — here's what changed and why" |
| Code patterns | CodeDB index | Symbol search, git history, cross-references across your codebase |
CodeDB: smarter code navigation
SageOx includes CodeDB — a semantic index of your codebase that Claude Code can query during sessions.
With CodeDB indexed, your agent can:
- Search by meaning — find functions and types by what they do, not just their name
- Trace git history — "when did this behavior change?" answered instantly
- Navigate cross-references — understand how code connects across modules
Without CodeDB, Claude Code relies on grep and file reads. With it, the agent has a map of your codebase.
Working with agent factories
SageOx works with any tool that orchestrates Claude Code sessions — including OpenClaw, agent factories, and custom orchestration setups.
The integration is the same regardless of how the session is launched:
- The orchestrator starts Claude Code in your repo
- Claude Code runs
ox agent primeon startup (configured via CLAUDE.md) - Team Context flows in automatically
- Session artifacts (decisions, patterns discovered) can be captured back
CLAUDE.md setup
When you run ox init, it adds the following to your repo's CLAUDE.md:
## SageOx
Run `ox agent prime` at session start, after /clear, and after /compact.This ensures context injection happens regardless of how the Claude Code session was started — terminal, IDE, API, or factory.
Multiple agents, shared context
When multiple agents (or multiple humans with agents) work on the same repo, they all receive the same Team Context. Decisions made in one session become available to all future sessions after they're committed.
This is the multiplayer loop: every session enriches the context that future sessions consume.
Capturing session knowledge
Beyond injecting context, SageOx can capture what happens during coding sessions:
The session transcript — your questions, the agent's reasoning, decisions made — gets committed to your repo's Ledger. Future sessions and coworkers can reference it.
Next steps
- Solo Quickstart — get started in 5 minutes
- Video Import — import screen recordings as team knowledge
- Web Recorder — record discussions from your browser