Contributors
Internal docs for people working on noClickOps — not just using it. If you're using noClickOps as a developer in your daily flow, start with the docs landing page instead.
What lives here
| Doc | When to read it |
|---|---|
| Target layout reference | Before changing any command that touches an Azure DevOps repo. Documents the Red Cross ADO + Azure layout noclickops targets, as observed in the wild. Re-verify after every Azure-engineer-side change. |
lib/service-v2.sh | When writing or modifying v2 commands. Documents the v2 discovery library — public API, discovery-vs-derivation-vs-override pattern, test shims, and the manual smoke procedure. |
What's not here yet (but will be, as the questions come up)
- How to add a new command — copy
bin/<existing>.sh, fill in theSCRIPT_*metadata block, add tests, commit per the PLAN workflow. - Release flow — semver bumps in
version.txt, branch + PR + merge, deploy-via-CI. - Slim install internals — the sparse-checkout + shallow-clone rationale, sparse-set evolution (v1.5.1 → v1.5.4).
- Testing — what
tests/run-all.shcovers, when to add a new test file. - Docs generator (
scripts/generate-docs.sh) — how SCRIPT_* metadata becomes per-command pages + how to extend it.
These will land here when someone needs them. Open an issue (or just ask) if you hit a contributor question this section doesn't answer.
Related reading
- AI Developer Workflow — the plan-based development convention this repo uses (
INVESTIGATE-*→PLAN-*→ implement → ship). Contributor docs describe the noclickops codebase; AI-developer docs describe how we change it. - The repo-root README, CLAUDE.md, and AGENTS.md — entry points for AI tooling that auto-loads project context.