plan
Align. Sequence the work. Work backwards from success.
mkdir -p ~/.claude/commands && curl -fsSL https://raw.githubusercontent.com/suboss87/FDEOps/HEAD/.claude/commands/plan.md -o ~/.claude/commands/plan.mdplan.md
Load `@fde` (`skills/fde/SKILL.md`). Stage: **plan**. If `fde resume` shows NO ENGAGEMENT: ask "What should we call this client?" once, then **you** run `fde resume --init`. Never tell the human to type the CLI. Run `fde resume` (fallback: `npx --yes fdeops resume`). Read `references/plan.md` and follow it. Sequence backwards from done, one change they can see at a time, who signs. Each Now item names `Kill if`. Do not invent a plan the record cannot support. Missing success criteria → `unknown - ask: <question>`. Confirm before any write to `.fde/`. Done when: order, the first visible changes, and “done” are on the record and the human agrees. Not for a one-line typo in an unbound file. Bound client work, a POC, eval, or go-live stays on `@fde`.
Engage. Interrogate the brief. Name who signs done.
Transfer operations. They operate it without you.
Capture the meeting. Notes into the record.
Diagnose. Check the brief is the real job.
Keeps the engagement record for client work. Use when they name a client or stakeholder. Use when they debrief a meeting or paste notes. Use when they ask what was agreed. Use when they run a POC, change the client's codebase, prove it on their staging, go live, or need evals before a model acts. Use when they prep a readout, when trust shifts, or they say @fde. Route and run the local fde CLI (or npx --yes fdeops). Never ask them to type commands. Not for ordinary code edits in an unbound repo.
Realize. Promised, measured, accepted.
Prepare the meeting. One page from the record.
Report the outcome. Promised, measured, accepted.