Guides
Task-oriented recipes, each with goal, prerequisites, steps, verification and links. Skills run in your assistant; pair-cli covers install, update and unattended runs; the Claude Code plugin is optional.
Each guide is one task: goal, prerequisites, numbered steps, how to verify, where to read more. For the flow these tasks belong to, read the Process Lifecycle first.
The three paths
Every step names the path that applies, with the same three labels everywhere in the docs:
- In your assistant: a skill you invoke as a slash command in Claude Code, Codex, Cursor, pi, opencode or any tool that reads Agent Skills. This is the default path for everything after install: bootstrap, planning, refinement, implementation, review.
- With pair-cli: a real
pair-clicommand in a shell:install,update,run,package,scaffold-kb,kb-*. Install, update and unattended runs live here; there is no CLI command for a skill. - With
/pair-assistant(optional, Claude Code): the one skill the Claude Code marketplace plugin ships. It runs the CLI for you and writes adoption files after showing the diff. Only listed where asking it is the thing to do; the other two paths never depend on it.
Set up
- Install via the Claude Code marketplace
- Adopt pair in an existing repository
- Join a project that already uses pair
- Update the Knowledge Base
- Configure another harness
Knowledge Base
- Customize the Knowledge Base for your team
- Customize templates
- Package-local rules in a monorepo
- Own and publish a Knowledge Base
Configure
- Card states, DoR/DoD and the process profile
- Classify stories and use risk tags
- Quality gates and the coverage ratchet
- Review identity and light auto-approval
- Assess security, cost and coupling
Deliver
- Deliver a card
- Autonomy and per-stage engine and model
- Cycle Hooks and Blocking Severities
- Deliver multiple cards
- Automate work via tags
- Measure delivery
Longer, learning-oriented walkthroughs are in Tutorials; every setting is in Adoption settings.