sdlc-vision

Solid

Externalize a tiered product north-star (vision -> strategy -> design -> architecture) into .sdlc/context/north-star.md, so goals are grounded in why you're building, not vibes. The opt-in "vision-first" on-ramp; drop-in (a thin project.md) stays the default. Use when the user runs /sdlc-vision or wants to start from a product vision.

AI & Automation 6 stars 1 forks Updated today MIT

Install

View on GitHub

Quality Score: 81/100

Stars 20%
28
Recency 20%
100
Frontmatter 20%
70
Documentation 15%
100
Issue Health 10%
80
License 10%
100
Description 5%
100

Skill Content

# sdlc-vision The opt-in **vision-first** on-ramp. Drop-in (a thin `project.md`, then run goals) stays the default; this is for starting **top-down from a product vision** so every goal is grounded in *why* you're building it. Same `.sdlc/` layout, same spine — just a thicker context layer. 1. **Scaffold the skeleton** if absent (skip-if-exists — never clobbers your edits): `python3 "${CLAUDE_SKILL_DIR}/../sdlc-init/scripts/sdlc_init.py" . --vision` It writes `.sdlc/context/north-star.md` with four tiers to fill. 2. **Draft from the repo first, then refine** — don't hand over a blank page. Read the README, the code structure, and recent git history, write a *first-pass draft* of each tier, then walk the user through refining it. Editing a draft beats filling a blank. Keep each short (direction, not a spec): - **Vision** — draft the change / audience you infer from the README and project; the user corrects. - **Strategy** — draft priorities + **non-goals** from recent commits / open issues; the user sets the real ones (the plan-review alignment gate uses the non-goals). - **Design** — draft the UX shape + the principles a change must respect; the user adjusts. - **Architecture** — draft the rules you develop by as a **numbered, checkable list** from the codebase (layering, dependency direction, boundaries) for the user to approve — plan-review **enforces** these (the stack stays in `project.md`). Write each tier into the file as you ...

Details

Author
swapnil-agrim
Repository
swapnil-agrim/loopsmith
Created
1 months ago
Last Updated
today
Language
Python
License
MIT

Integrates with

Similar Skills

Semantically similar based on skill content — not just same category