pm-prioritizelisted

Re-runnable prioritization engine for the feature backlog. Works at ANY point - after the feature list is created, after the MVP cut, or any time later when priorities shift. Takes flexible input - align to a roadmap, follow a directive you give ("X first", "revenue first"), apply a prioritization lens, or let the skill propose a basis with reasoning. Always reconciles with the dependency map (cannot rank a feature ahead of its blocker), is non-destructive (proposes, you confirm before it writes), and explains the rationale and trade-offs. The canonical prioritization logic that pm-features-list offers for the first-time scoring.
ljucask/pureinn-product-development · ★ 1 · AI & Automation · score 77

Install: claude install-skill ljucask/pureinn-product-development

# PM - Prioritize (Backlog Prioritization Engine) ## Agent mode (`--agent`) Supports `--agent`: runs autonomously in a subagent, drafts the artifact from existing inputs, and returns a short summary + coverage note. - **No flag** → interactive (default); if inputs are heavy, offer agent mode. - **`--agent`** → obey. First check inputs are complete. Anything missing: do NOT invent it - mark `[ASSUMED - what/why]` in the output and summary. Never hallucinate to fill a gap. - **Review required:** the artifact contains commitments - after drafting, require the user's review before finalizing; do not close decisions autonomously. --- ## What this skill does Prioritizes (or re-prioritizes) the feature backlog at **any point in the lifecycle**, on a basis you choose or one it proposes. It is the canonical prioritization engine - `pm-features-list` offers it for the first-time scoring, and you re-run it standalone whenever priorities shift (new market signal, investor input, strategy change, a fresh roadmap). It is deliberately **agile, not waterfall** (per the Adaptive-execution standard in CLAUDE.md): it does not auto-impose a default order - it **offers** a basis (and proposes one with reasoning when you are unsure), reconciles against dependencies, shows the rationale and trade-offs, and only writes after you confirm. Re-running it iterates the order; it never destroys the KANO/V×C scores or the feature data. **Produces:** an updated priority / sequence on `features/feat