radin-planlisted

Write a step-by-step implementation plan for one backlog entry, without touching code. Takes a task scope — a title/keyword — instead of the whole backlog. Judges whether the scope is broad enough to split into multiple independent plans, confirms any split with you directly, then writes one plan file per resulting sub-task, reviews it with thermo-nuclear and ponytail-review before handing it off, and appends a `**Plan:**` pointer back to the entry. Use for /radin-plan, "plan this backlog entry", "write a plan for X before we execute it". radin-execute delegates to a planning sub-agent that invokes this skill, non-interactively, for any entry it judges too complex to implement without a plan.
shortcuts/radin · ★ 3 · AI & Automation · score 71

Install: claude install-skill shortcuts/radin

# Plan a Backlog Entry Turn one `BACKLOG.md` entry into one or more concrete implementation plans, without writing any code. This runs inline in whichever context invokes it, so any split judgment or open question surfaces in that conversation. When the invoking context cannot reach the user (e.g. radin-execute's planning sub-agent), the caller says so and this skill's questions resolve to their non-destructive defaults: no split, no overwrite. ## Step 1: Resolve project namespace, locate BACKLOG_FILE All backlog reads/writes go through the shared CLI at `$HOME/.claude/.radin/lib/radin-backlog.sh` — never hand-edit `BACKLOG.md` or compute its path yourself. Get the paths (also creates `$NAMESPACE_DIR/state`, `plans/`, `reviews/`): ```bash bash "$HOME/.claude/.radin/lib/radin-backlog.sh" env ``` Read `REPO_ROOT`, `NAMESPACE_DIR`, `BACKLOG_FILE` from its output. Re-run this line in any later Bash call before using these variables. ## Step 2: Resolve the task scope Match the caller's scope against entries with the CLI — it prints one `line_start<TAB>line_end<TAB>title` line per matching entry (exact title match first, else case-insensitive substring): ```bash bash "$HOME/.claude/.radin/lib/radin-backlog.sh" find "<scope title/keyword>" ``` - **Exactly one match**: use it. - **Multiple candidate matches**: list them and ask which one. Invoked non-interactively: don't guess — report the candidates and stop; the caller marks the task blocked for the user. - **No match*