@polymorphism-tech/morph-spec 4.21.0 → 5.0.1

package/framework/skills/level-1-workflows/morph-phase-codebase-analysis/SKILL.md

@@ -1,269 +0,0 @@
----
-name: morph-phase-codebase-analysis
-description: MORPH-SPEC Design sub-phase that analyzes existing codebase and database schema, producing schema-analysis.md with real column names, types, relationships, and field mismatches. Use at the start of Design phase before generating contracts.cs to prevent incorrect field names or types.
-user-invocable: false
-allowed-tools: Read, Write, Edit, Bash, Glob, Grep, AskUserQuestion
-cliVersion: "4.21.0"
----
-
-# MORPH Codebase Analysis - Design Sub-phase
-
-> INTERNAL: Automation skill for schema analysis step within Phase 2 (Design).
-> Called by `phase-design.md` Step 2. Not a standalone user command.
-
-Automates existing code analysis to generate `schema-analysis.md` with real database/code data.
-
-## Prerequisites
-
-- [ ] Proposal phase completed
-- [ ] Feature has approved `proposal.md`
-- [ ] Project context loaded (stack, agents)
-
-## Recommended Tools
-
-> **Ref:** `framework/skills/level-0-meta/morph-tool-usage-guide/SKILL.md` for complete guide.
-> **Ref:** `framework/standards/integration/mcp/mcp-tools.md` for MCP reference.
-
-| Action | Tool | Alternative |
-|------|------------|-------------|
-| List database tables | **Bash** `pg_dump $(neonctl connection-string)` or `psql <DB_URL>` | **Grep** `.from(` in code |
-| Get table schema | **Bash** `psql <DB_URL> -c "SELECT ... FROM information_schema.columns"` | **Read** type definitions |
-| Get FK relationships | **Bash** `psql <DB_URL> -c "SELECT ... FROM information_schema.table_constraints"` | **Grep** JOIN/FK patterns |
-| Get RLS policies | **Bash** `psql <DB_URL> -c "SELECT ... FROM pg_policies"` | **Read** policy files |
-| Find queries in code | **Grep** `\.from\(` in `*.ts,*.tsx,*.js,*.cs` | — |
-| Find type definitions | **Glob** `src/**/types/**/*.ts` or `**/Entities/**/*.cs` | — |
-| Read found files | **Read** each file | — |
-| Complex multi-file analysis | **Task** (subagent Explore) | Read individual |
-| Generate schema-analysis.md | **Write** using template | **Bash** template render |
-
-**Tools for this phase:** Neon CLI (`neonctl`) / psql (schema — **PRIORITY**), Generic database CLIs.
-
----
-
-## Automated Workflow
-
-### Step 1: Detect Analysis Method
-
-**IMPORTANT:** Do not assume the MCP is available and accessible.
-Verify first with a test call before attempting to use it.
-
-```
-0. **Database connectivity check:** Run `neonctl connection-string` to verify Neon CLI is configured.
-   → IF Neon CLI is not configured → warn user: "Neon CLI not configured — run `neonctl auth`."
-   → Ask user via AskUserQuestion: configure Neon or fall back to Method B.
-
-1. IF Neon/PostgreSQL database detected (DATABASE_URL env var exists OR config.json stack includes database):
-   → Run `neonctl connection-string` to get the DB URL
-   → IF running (exit 0): extract DB URL from output → Use Method A (direct CLI/psql)
-   → IF not running (exit non-zero):
-      → Log: "Neon database not accessible — run `neonctl auth`"
-      → Ask user via AskUserQuestion: configure it or fall back to Method B
-
-2. IF no Neon/PostgreSQL database detected:
-   → Use Method B (static code analysis)
-```
-
-**Signs of unavailable Neon database:**
-- `neonctl connection-string` exits with non-zero code
-- Neon CLI not authenticated or project not configured
-- Project uses database other than Neon (CloudSQL, RDS, Azure SQL, etc.)
-
-### Step 2A: CLI Method (Preferred — Neon CLI)
-
-**Tools:** Bash (Neon CLI + psql)
-
-```bash
-# 0. Get DB URL
-neonctl connection-string
-# → Returns connection string, e.g.: postgresql://user:pass@ep-xxx.us-east-2.aws.neon.tech/neondb
-
-# 1. List all tables
-pg_dump $(neonctl connection-string) --schema public 2>/dev/null | grep "CREATE TABLE"
-
-# 2. For each feature-relevant table:
-
-# 2a. Complete schema
-psql $DB_URL -c "SELECT column_name, data_type, is_nullable, column_default \
-  FROM information_schema.columns WHERE table_name = 'TABLE_NAME' ORDER BY ordinal_position"
-
-# 2b. Relationships (FK)
-psql $DB_URL -c "SELECT tc.table_name, kcu.column_name, ccu.table_name AS foreign_table, \
-  ccu.column_name AS foreign_column FROM information_schema.table_constraints tc \
-  JOIN information_schema.key_column_usage kcu ON tc.constraint_name = kcu.constraint_name \
-  JOIN information_schema.constraint_column_usage ccu ON tc.constraint_name = ccu.constraint_name \
-  WHERE tc.constraint_type = 'FOREIGN KEY' AND tc.table_name = 'TABLE_NAME'"
-
-# 2c. Indexes
-psql $DB_URL -c "SELECT indexname, indexdef FROM pg_indexes WHERE tablename = 'TABLE_NAME'"
-
-# 2d. RLS policies (security)
-psql $DB_URL -c "SELECT policyname, cmd, qual FROM pg_policies WHERE tablename = 'TABLE_NAME'"
-```
-
-### Step 2B: Static Analysis Method (Fallback)
-
-**Tools:** Grep, Glob, Read, Task (subagent for large projects)
-
-**Phase B1: Find Queries**
-
-```
-Grep: "\.from\(|\.select\(|SELECT |context\.|dbContext\.|DbSet<"
-Type: ts,tsx,js,jsx,cs
-Output: files_with_matches
-```
-
-**Phase B2: Read Query Files**
-
-For each found file, use **Read** to extract:
-- Table names: `from('leads')`, `DbSet<Lead>`, `FROM leads`
-- Column names: `.select('fullname, phonenumber')`, `l.FullName`
-- Data types: TypeScript interfaces, C# DTOs
-- Relationships: JOIN clauses, navigation properties
-
-**Phase B3: Find Type Definitions**
-
-```
-# TypeScript/JavaScript:
-Glob: "src/**/types/**/*.ts"
-Glob: "src/**/*.d.ts"
-Glob: "src/**/interfaces/*.ts"
-
-# .NET:
-Glob: "**/*Dto.cs"
-Glob: "**/Entities/**/*.cs"
-Glob: "**/Models/**/*.cs"
-```
-
-**Phase B4: When to use Task (subagent)**
-
-Use Task subagent **ONLY** when:
-- Project has 20+ query files
-- Multiple data access patterns (Neon + EF Core + raw SQL)
-- Precise cross-context analysis across many files
-
-Do not use Task subagent when:
-- Project has < 10 query files (direct Read is faster)
-- Data access pattern is uniform (only Neon OR only EF Core)
-
-### Step 2B-Format: Standardized Format for Static Analysis
-
-**IMPORTANT:** When using Method B (static analysis), **always use the official template**.
-This ensures `contracts.cs` can be generated mechanically, without manual interpretation.
-
-The template is at `framework/templates/docs/schema-analysis.md`. Use it via:
-
-```bash
-morph-spec template render docs/schema-analysis \
-  .morph/features/{feature}/2-design/schema-analysis.md \
-  '{
-    "FEATURE_NAME": "{feature}",
-    "FEATURE_NAME_TITLE": "{Feature Display Name}",
-    "DATE": "YYYY-MM-DD",
-    "AUTHOR": "Claude Code (Sonnet 4.6)",
-    "method": "Static Code Analysis",
-    "mcpUsed": "No",
-    "tableCount": N,
-    "fileCount": N,
-    "tables": [
-      {
-        "name": "table_name",
-        "codeSource": "path/to/file.ts",
-        "columns": [
-          { "name": "column_name", "type": "string", "nullable": false, "notes": "from SELECT query" }
-        ],
-        "relationships": [{ "description": "belongs to users via user_id" }],
-        "indexes": [{ "description": "idx_leads_created_at (created_at DESC)" }]
-      }
-    ],
-    "fieldMismatches": [
-      { "description": "Code uses user.name but DB column is users.fullname" }
-    ],
-    "typeMismatches": [],
-    "relationshipIssues": [],
-    "dtoRecommendations": [
-      {
-        "tableName": "leads",
-        "dtoName": "LeadDto",
-        "fields": [
-          { "type": "Guid", "name": "Id", "nullable": false, "comment": "PK" },
-          { "type": "string", "name": "FullName", "nullable": false, "comment": "leads.fullname" }
-        ]
-      }
-    ]
-  }'
-```
-
-**If template render is not available**, use **Write** tool to create the file directly
-following the structure above. The table and mismatch format is MANDATORY — without standard format
-it's not possible to generate contracts.cs mechanically.
-
-### Step 3: Map Inconsistencies
-
-Create a map of:
-
-| Frontend/Code | Database | Type | Problem |
-|----------------|----------------|------|----------|
-| user.name | users.fullname | string | MISMATCH |
-| lead.phone | leads.phonenumber | string | MISMATCH |
-| order.metadata | orders.metadata | JSONB | Type complex |
-
-### Step 4: Generate `schema-analysis.md`
-
-Use the template at `framework/templates/docs/schema-analysis.md`:
-
-```bash
-morph-spec template render docs/schema-analysis \
-  .morph/features/{feature-name}/2-design/schema-analysis.md \
-  '{ "FEATURE_NAME": "{feature-name}", "DATE": "..." }'
-```
-
-OR use **Write** tool to create directly with collected data.
-
-### Step 5: Update State
-
-```bash
-morph-spec state mark-output {feature-name} schema-analysis
-```
-
----
-
-## Outputs
-
-- `.morph/features/{feature}/2-design/schema-analysis.md`
-- State updated with `schemaAnalysis` output
-
-## CHECKPOINT
-
-Before proceeding to contracts.cs, present results and ask for confirmation using `AskUserQuestion`:
-
-```json
-{
-  "questions": [
-    {
-      "header": "Schema Review",
-      "question": "Schema analysis complete: {N} tables, {N} field mismatches, {N} type mismatches, {N} relationships. Is the analysis correct?",
-      "multiSelect": false,
-      "options": [
-        { "label": "Correct", "description": "Schema analysis matches the real database" },
-        { "label": "Has issues", "description": "I see problems — let me explain" }
-      ]
-    },
-    {
-      "header": "Contracts",
-      "question": "May I generate contracts.cs based on this real schema?",
-      "multiSelect": false,
-      "options": [
-        { "label": "Generate contracts", "description": "Proceed to contracts.cs using the analyzed schema" },
-        { "label": "Wait", "description": "I want to review schema-analysis.md first" }
-      ]
-    }
-  ]
-}
-```
-
-**If "Has issues"** → ask what's wrong, fix schema-analysis.md, re-present checkpoint.
-**If "Correct" + "Generate contracts"** → proceed to contracts generation.
-
----
-
-*MORPH-SPEC by Polymorphism Tech*