Resources for working with a coding agent.
In-depth guides to make Claude Code or Codex work with method, a glossary of Treelab’s words and what we do about security. Every page points to the real commands.
Guides
- Git worktrees: running several coding agents in parallel
One git worktree per task: each agent gets its own folder and branch, on the same repository.
- The lot method: one measurable objective per agent session
Scope, split into lots, prove: one measurable objective per Claude Code or Codex session.
- Staying in control of a coding agent, without watching it all the time
Decide at the right moments, let the agent do the rest, require evidence before each delivery.
Reference
- Glossary
Treelab’s and coding agents’ words, defined in one or two sentences.
- Security: what Treelab protects, and what it does not claim to protect
Guardrails, their limits, the data that leaves your Mac and how to report a vulnerability.
Also on this site
- Features
Every lab command, with what it prevents.
- Guide
Install, start, move forward, finish: every command in detail.
- What’s new
What changes in each published version.
Try it on your project
One command in Terminal to buy and install. One payment, lifetime updates.
curl -fsS https://treelab.sh/install | /bin/bashTreelab is an independent product, with no affiliation to OpenAI, Anthropic or cmux.