sr:save

sr:save

Goal: distil the entire project into compact, re-loadable context files. A future session should be able to load just _context/ and have full project understanding.

---

Step 0 — Identify Project

AskUserQuestion: "Which project to save context for? (slug)"

Read all available files in projects/{slug}/. Run python .claude/scripts/srs_validator.py --dir projects/{slug}/srs/ --stats and use its counts for quality_standards.md / session-notes.md below instead of recounting by hand.

---

Step 1 — Write Context Files

Write all files under projects/{slug}/_context/:

vision.md

# Vision — {Project Name}

## Problem Statement
{what problem does this system solve? 2–3 sentences from spec §1}

## Solution Summary
{what the system does, for whom, and the core value proposition}

## Success Metrics
{how we measure success — from spec §1 or brainstorm}

## Constraints
{top 3 hard constraints — tech, timeline, compliance}

features.md

# Features — {Project Name}

## IN Scope (v1.0)
| Feature | Priority | Key FRs | Notes |
|---------|----------|---------|-------|
{every confirmed in-scope feature, one row each}

## OUT of Scope (v1.0 — Deferred)
| Feature | Reason | Target version |
|---------|--------|---------------|
{every explicitly excluded feature}

tech_stack.md

# Tech Stack — {Project Name}

## Confirmed Stack
| Layer | Technology | Constraint type |
|-------|-----------|----------------|
{language, framework, DB, cloud, CI/CD — from spec §5}

## Integration Points
| System | Protocol | Direction | Auth |
|--------|----------|-----------|------|
{every external system — from SRS §3.1.3}

## Compliance Requirements
{applicable regulations + specific technical obligations}

glossary.md

# Glossary — {Project Name}

Copy of Appendix A from the SRS. Every term, acronym, abbreviation.
Include: domain terms | project-specific terms | role names | system names