db-mcp 1.1.0 → 2.0.0

package/test-server/test-tool-groups/test-text-basic.md

@@ -0,0 +1,177 @@
+# db-mcp Tool Group Testing: [text-basic]
+
+> [!IMPORTANT]
+> **Do not track progress in this file.** Track your test progress, coverage matrix, and findings in your internal task tracking system (artifact). However, you SHOULD edit this file to fix any factual errors, broken code, or incorrect assertions in the test prompts.
+> If there is nothing to fix, don't update UNRELEASED.md.
+> We're currently testing Native mode.
+
+## WASM Mode
+
+> When testing against a **WASM backend** (`sqlite-wasm` / sql.js): All 11 tools in this basic text suite are fully WASM-compatible. No items to skip or adjust.
+
+**Step 1:** Confirm you read the server help content sourced from `C:\Users\chris\Desktop\db-mcp\src\constants\server-instructions\gotchas.md` using `view_file` (not grep or search) — to understand documented behaviors, edge cases, and response structures for this tool group.
+
+**Step 2:** Please conduct an exhaustive test of the **text-basic** tool group specified in the group-specific checklist below using live MCP server tool calls directly — not scripts/terminal.
+
+**Step 3:** The agent should update C:\Users\chris\Desktop\db-mcp\UNRELEASED.md with any/all changes/fixes.
+
+**Note** If temp tables are present from a previous test pass, it's because the database is locked. Ignore them.
+
+## Reporting Format
+
+- ❌ Fail: Tool errors or produces incorrect results (include error message)
+- ⚠️ Issue: Unexpected behavior or improvement opportunity
+- 📦 Payload: Unnecessarily large response that should be optimized.
+
+## Test Database Schema
+
+The test database (test-server/test.db) contains these tables with JSON-relevant columns:
+
+| Table             | Rows | Columns                                                                       | JSON Columns                                                                              |
+| ----------------- | ---- | ----------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------- |
+| test_products     | 16   | id, name, description, price, category, created_at                            | —                                                                                         |
+| test_orders       | 20   | id, product_id (FK), customer_name, quantity, total_price, order_date, status | —                                                                                         |
+| test_jsonb_docs   | 6    | id, doc, metadata, tags, created_at                                           | **doc**, **metadata** (nested), **tags** (array)                                          |
+| test_articles     | 8    | id, title, body, author, category, published_at                               | —                                                                                         |
+| test_users        | 9    | id, username, email, phone, bio, created_at                                   | —                                                                                         |
+| test_measurements | 200  | id, sensor_id, temperature, humidity, pressure, measured_at                   | —                                                                                         |
+| test_embeddings   | 20   | id, content, category, embedding                                              | **embedding** (8-dim float array); category values: database, fitness, food, tech, travel |
+| test_locations    | 15   | id, name, city, latitude, longitude, type                                     | —                                                                                         |
+| test_categories   | 17   | id, name, path, level                                                         | —                                                                                         |
+| test_events       | 100  | id, event_type, user_id (INT, 8 values), payload, event_date                  | **payload** (JSON)                                                                        |
+
+## Testing Requirements
+
+> [!CAUTION]
+> **Zero tolerance for raw MCP errors.** ANY response that is a raw MCP error (e.g., `-32602`, `isError: true`, no `success` field) is a **bug that must be reported and fixed** — never an acceptable design choice, SDK limitation, or expected behavior. If you see one, report it as ❌ immediately. Do not rationalize it as "the SDK rejecting at the boundary" or "by design for range-constrained params." The handler MUST catch it.
+
+1. Use existing `test_*` tables for read operations.
+2. Test each tool with realistic inputs based on the schema above.
+3. Report all failures, unexpected behaviors, improvement opportunities, or unnecessarily large payloads.
+4. **Error path testing**: For **every** tool, test at least **two** invalid inputs: (a) a domain error and (b) a **Zod validation error** (call the tool with `{}` empty params). Both must return a **structured handler error** (`{success: false, error: "..."}`) — NOT a raw MCP error frame.
+5. **Output schema testing**: For **every** tool that has an `outputSchema`, confirm that at least one valid happy-path call returns a structured JSON response — NOT a raw MCP `-32602` "output schema" error. Output schema mismatches produce the same `-32602` code as input errors but are only caught with valid inputs.
+6. **Code Over Docs**: Fix the handler code if standards (Structured Errors/Zod) are violated. Do NOT change docs/prompts to accommodate broken code.
+
+## Structured Error Response Pattern
+
+All tools should return errors as structured objects instead of throwing. The expected pattern:
+
+```json
+{ "success": false, "error": "Human-readable error message" }
+```
+
+### Handler Error vs MCP Error — How to Distinguish
+
+| Type                 | Source                                                             | What you see                                                                                                          | Verdict            |
+| -------------------- | ------------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------------------- | ------------------ |
+| **Handler error** ✅ | Handler catches error and returns `{success: false, error: "..."}` | Parseable JSON object with `success` and `error` fields                                                               | Correct            |
+| **MCP error** ❌     | Uncaught throw propagates to MCP framework                         | Raw text error string, often prefixed with `Error:`, wrapped in an `isError: true` content block — no `success` field | Bug — report as ❌ |
+
+### Zod Validation Errors
+
+Calling a tool with wrong parameter types or missing required fields triggers a Zod validation error. If the handler has no outer `try/catch`, this surfaces as a raw MCP error (often `-32602`). Test every tool with `{}` (empty params) if it has required parameters — the response must be a handler error, not an MCP error.
+
+**Fix:** Remove ALL `.min(N)` / `.max(N)` refinements from the schema and validate inside the handler instead.
+
+### Wrong-Type Numeric Parameter Coercion
+
+For every tool with optional numeric parameters (e.g., `limit`), call the tool with `param: "abc"` (string instead of number). The tool must NOT return a raw MCP `-32602` error.
+
+### Output Schema Validation Errors
+
+The MCP SDK enforces `additionalProperties: false` on **output** schemas. If a handler returns fields not declared in its output schema, the SDK rejects the response with a raw `-32602` error.
+
+**How to detect:** If a tool call with **correct, valid inputs** returns a raw MCP `-32602` mentioning "does not match the tool's output schema" or "additional properties", report as ❌ with both the tool name and the missing field(s).
+
+### Error Consistency Audit
+
+1. **Throw-vs-return**: If a tool throws a raw error instead of returning `{success: false}`, report as ❌.
+2. **Error field name**: All `{success: false}` responses should use `error` as the field name.
+3. **Zod validation leaks**: If calling a tool with missing required field produces a raw MCP `-32602` error instead of a structured response, report as ❌.
+4. **Output schema leaks**: If calling a tool with valid inputs produces a raw MCP `-32602` mentioning "output schema", report as ❌.
+
+---
+
+## Group Focus: text-basic
+
+> **Instructions**: Execute every numbered checklist item with the exact inputs shown. Compare responses against the expected results. Report any deviation.
+
+### text-basic Group Tools (11)
+
+1. sqlite_regex_extract
+2. sqlite_regex_match
+3. sqlite_text_split
+4. sqlite_text_concat
+5. sqlite_text_replace
+6. sqlite_text_trim
+7. sqlite_text_case
+8. sqlite_text_substring
+9. sqlite_text_validate
+10. sqlite_text_normalize
+11. sqlite_execute_code
+
+**Test data reference:**
+
+- `test_users` (9 rows): Emails include `@example.com`, `@company.org`, `@gmail.com`, etc. One user (`testuser`) has `test.user@gmail.com`. Phone formats: `+1-555-0101`, `+44-20-7123-4567`, `+82-2-1234-5678`
+- `test_products` row 16: `name = 'Café Décor Light'` — has accented characters for `strip_accents` testing
+
+**Checklist:**
+
+1. `sqlite_regex_match({table: "test_users", column: "email", pattern: "@gmail\\.com$"})` → at least 1 result (`test.user@gmail.com`)
+2. `sqlite_regex_extract({table: "test_users", column: "email", pattern: "@([^.]+)\\.", groupIndex: 1})` → extract domain parts (example, company, startup, etc.)
+3. `sqlite_text_validate({table: "test_users", column: "email", pattern: "email"})` → all 9 rows should be valid emails
+4. `sqlite_text_validate({table: "test_users", column: "phone", pattern: "phone"})` → report valid/invalid counts (one user has NULL phone)
+5. `sqlite_text_case({table: "test_users", column: "username", mode: "upper"})` → all usernames uppercased
+6. `sqlite_text_normalize({table: "test_products", column: "name", mode: "strip_accents"})` → `Café Décor Light` becomes `Cafe Decor Light`
+7. `sqlite_text_split({table: "test_users", column: "email", delimiter: "@"})` → each email split into local + domain parts
+8. `sqlite_text_concat({table: "test_users", columns: ["username", "email"], separator: " - "})` → concatenated strings
+9. `sqlite_text_replace({table: "test_users", column: "email", searchPattern: "@example.com", replaceWith: "@test.org", whereClause: "email LIKE '%@example.com'"})` → 1 row affected (write operation — revert with `searchPattern: "@test.org", replaceWith: "@example.com", whereClause: "email LIKE '%@test.org'"` afterward)
+10. `sqlite_text_trim({table: "test_users", column: "bio"})` → trimmed bios
+11. `sqlite_text_substring({table: "test_users", column: "username", start: 1, length: 4})` → first 4 chars of each username
+
+**Code mode testing:**
+
+12. `sqlite_execute_code({code: "const result = await sqlite.text.regexMatch({table: 'test_users', column: 'email', pattern: '@gmail\\\\.com$'}); return result;"})` → at least 1 result
+
+**Error path testing:**
+
+🔴 13. `sqlite_regex_match({table: "nonexistent_table_xyz", column: "x", pattern: "."})` → structured error
+
+**Zod validation sweep** — call each tool with `{}` (empty params). Must return handler error, NOT raw MCP error:
+
+🔴 14. `sqlite_regex_extract({})` → handler error
+🔴 15. `sqlite_regex_match({})` → handler error
+🔴 16. `sqlite_text_split({})` → handler error
+🔴 17. `sqlite_text_concat({})` → handler error
+🔴 18. `sqlite_text_replace({})` → handler error
+🔴 19. `sqlite_text_trim({})` → handler error
+🔴 20. `sqlite_text_case({})` → handler error
+🔴 21. `sqlite_text_substring({})` → handler error
+🔴 22. `sqlite_text_normalize({})` → handler error
+🔴 23. `sqlite_text_validate({})` → handler error
+🔴 24. `sqlite_execute_code({})` → handler error
+
+---
+
+## Post-Test Procedures
+
+### Reporting Rules
+
+- Use ✅ only in inline notes during testing; omit from Final Summary
+- Do not mention what already works well or issues already documented in help resources and runtime hints
+
+### After Testing
+
+1. **Triage findings**: If issues were found, create an implementation plan, making sure they are consistent with working patterns in other tools/tool groups. If the plan requires no user decisions, proceed directly to implementation
+2. **Scope of fixes** includes corrections to any of:
+   - Handler code
+   - `src/constants/server-instructions/*.md` (per-group help files) — run `npm run generate:instructions` after editing to regenerate `server-instructions.ts`
+   - Test database (`test-server/test.db`)
+   - This prompt
+
+### After Implementation
+
+3. **Validate**: Instruct the user to run the test suite (Vitest/Playwright), lint, and typecheck. Do NOT run them yourself.
+4. **Commit**: Stage and commit all changes — do NOT push
+5. **Live re-test**: Test fixes with direct MCP tool calls. I will have already rebuilt and restarted the server.
+6. **Final summary**: If no issues found, provide the final summary after testing. If issues were fixed, provide the summary after live MCP re-testing confirms fixes are working.

package/test-server/test-tool-groups/test-transactions.md

@@ -0,0 +1,179 @@
+# db-mcp Tool Group Testing: [transactions]
+
+> [!IMPORTANT]
+> **Do not track progress in this file.** Track your test progress, coverage matrix, and findings in your internal task tracking system (artifact). However, you SHOULD edit this file to fix any factual errors, broken code, or incorrect assertions in the test prompts.
+> If there is nothing to fix, don't update UNRELEASED.md.
+> We're currently testing Native mode.
+
+## WASM Mode
+
+> [!CAUTION]
+> **Skip this entire prompt in WASM mode.** All 8 transaction tools are `[NATIVE ONLY]` — they are not registered in the WASM adapter. Direct MCP calls will fail with "unknown tool". Use `test-tool-group-codemode-wasm-degradation.md` to verify this behavior.
+
+**Step 1:** Confirm you read the server help content sourced from `C:\Users\chris\Desktop\db-mcp\src\constants\server-instructions\gotchas.md` using `view_file` (not grep or search) — to understand documented behaviors, edge cases, and response structures for this tool group.
+
+**Step 2:** Please conduct an exhaustive test of the **transactions** tool group specified in the group-specific checklist below using live MCP server tool calls directly — not scripts/terminal.
+
+**Step 3:** The agent should update C:\Users\chris\Desktop\db-mcp\UNRELEASED.md with any/all changes/fixes.
+
+**Note** If temp tables are present from a previous test pass, it's because the database is locked. Ignore them.
+
+## Reporting Format
+
+- ❌ Fail: Tool errors or produces incorrect results (include error message)
+- ⚠️ Issue: Unexpected behavior or improvement opportunity
+- 📦 Payload: Unnecessarily large response that should be optimized — **blocking, equally important as ❌ bugs**. Report the response size in KB and suggest a concrete optimization.
+
+## Test Database Schema
+
+| Table             | Rows | Columns                                                                       | JSON Columns                                                                              |
+| ----------------- | ---- | ----------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------- |
+| test_products     | 16   | id, name, description, price, category, created_at                            | —                                                                                         |
+| test_orders       | 20   | id, product_id (FK), customer_name, quantity, total_price, order_date, status | —                                                                                         |
+| test_jsonb_docs   | 6    | id, doc, metadata, tags, created_at                                           | **doc**, **metadata** (nested), **tags** (array)                                          |
+| test_articles     | 8    | id, title, body, author, category, published_at                               | —                                                                                         |
+| test_users        | 9    | id, username, email, phone, bio, created_at                                   | —                                                                                         |
+| test_measurements | 200  | id, sensor_id, temperature, humidity, pressure, measured_at                   | —                                                                                         |
+| test_embeddings   | 20   | id, content, category, embedding                                              | **embedding** (8-dim float array); category values: database, fitness, food, tech, travel |
+| test_locations    | 15   | id, name, city, latitude, longitude, type                                     | —                                                                                         |
+| test_categories   | 17   | id, name, path, level                                                         | —                                                                                         |
+| test_events       | 100  | id, event_type, user_id (INT, 8 values), payload, event_date                  | **payload** (JSON)                                                                        |
+
+> **Note:** When testing `sqlite_execute_code`, do **not** pass `readonly: true` unless specifically testing read-only filtering.
+
+## Testing Requirements
+
+> [!CAUTION]
+> **Zero tolerance for raw MCP errors.** ANY response that is a raw MCP error (e.g., `-32602`, `isError: true`, no `success` field) is a **bug that must be reported and fixed**.
+
+1. Use existing `test_*` tables for read operations
+2. Create temporary tables with `temp_*` prefix for write operations
+3. Report all failures, unexpected behaviors, or unnecessarily large payloads
+4. Do not mention what already works well or issues documented in help resources
+5. **Error path testing**: For **every** tool, test (a) domain error and (b) Zod validation error (`{}`). Both must return `{success: false, error: "..."}`.
+6. **Output schema testing**: For tools with `outputSchema`, confirm valid calls return structured JSON.
+7. **Deterministic checklist first**: Complete ALL items before freeform exploration.
+8. **Code Over Docs**: Fix the handler code if standards (Structured Errors/Zod) are violated. Do NOT change docs/prompts to accommodate broken code.
+
+## Structured Error Response Pattern
+
+```json
+{ "success": false, "error": "Human-readable error message" }
+```
+
+| Type                 | Source                                                             | What you see                                                | Verdict            |
+| -------------------- | ------------------------------------------------------------------ | ----------------------------------------------------------- | ------------------ |
+| **Handler error** ✅ | Handler catches error and returns `{success: false, error: "..."}` | Parseable JSON object with `success` and `error` fields     | Correct            |
+| **MCP error** ❌     | Uncaught throw propagates to MCP framework                         | Raw text error string, `isError: true` — no `success` field | Bug — report as ❌ |
+
+### Zod Validation Errors
+
+**Zod refinement leak pattern:** `.partial()` does NOT strip `.min(N)` / `.max(N)` refinements. **Fix:** Remove refinements from schema, validate inside handler.
+
+- Raw MCP error (no `success` field) → report as ❌
+- `{success: false, error: "..."}` → correct
+- Successful response for invalid input → report as ⚠️
+
+### Output Schema Validation Errors
+
+If valid inputs return raw MCP `-32602` mentioning "output schema", report as ❌.
+
+### Error Consistency Audit
+
+1. Raw error instead of `{success: false}` → ❌
+2. Must use `error` field name
+3. Orphaned/inline output schemas → ⚠️
+
+### Split Schema Pattern Verification
+
+Verify parameter visibility and alias acceptance.
+
+## Cleanup Conventions
+
+- **Temporary tables**: Prefix with `temp_`
+- If DROP fails due to database lock, move on.
+
+---
+
+## Group Focus: transactions
+
+> **Instructions**: Execute every numbered checklist item with the exact inputs shown. Compare responses against the expected results. Report any deviation.
+
+> **Note:** All 8 transaction tools are **`[NATIVE ONLY]`** — they are not available in WASM mode. Transaction operations require the `write` OAuth scope.
+
+### Built-in Tools (3)
+
+1. server_info
+2. server_health
+3. list_adapters
+
+### transactions Group Tools (8) `[NATIVE ONLY]`
+
+4. sqlite_transaction_begin
+5. sqlite_transaction_status
+6. sqlite_transaction_commit
+7. sqlite_transaction_rollback
+8. sqlite_transaction_savepoint
+9. sqlite_transaction_release
+10. sqlite_transaction_rollback_to
+11. sqlite_transaction_execute
+12. sqlite_execute_code
+
+**Checklist — Full Lifecycle:**
+
+1. `sqlite_transaction_status` → `{status: "none", active: false}` (no active transaction)
+2. `sqlite_transaction_begin` → capture transaction ID
+3. `sqlite_transaction_status` → `{status: "active", active: true}`
+4. `sqlite_transaction_rollback` → rollback entire transaction, verify success
+5. `sqlite_transaction_status` → `{status: "none", active: false}` (back to none)
+6. `sqlite_transaction_begin` → start a new transaction for savepoint tests
+7. `sqlite_transaction_savepoint({name: "sp1"})` → success
+8. `sqlite_transaction_rollback_to({name: "sp1"})` → success
+9. `sqlite_transaction_release({name: "sp1"})` → success (released savepoints cannot be rolled back to)
+10. `sqlite_transaction_commit` → success
+11. `sqlite_transaction_execute({statements: ["SELECT 1 AS test", "SELECT 2 AS test2"]})` → success with 2 statements executed
+
+**Code mode testing:**
+
+12. `sqlite_execute_code({code: "const result = await sqlite.transactions.status(); return result;"})` → transaction status
+13. `sqlite_execute_code({code: "const result = await sqlite.transactions.execute({statements: ['SELECT 1 AS test']}); return result;"})` → success
+
+**Error path testing:**
+
+🔴 14. `sqlite_transaction_execute({statements: ["INSERT INTO nonexistent_table VALUES (1)"]})` → structured error with rollback info
+🔴 15. `sqlite_transaction_execute({statements: []})` → report behavior for empty array
+
+**Zod validation sweep** — call each tool with `{}` (empty params). Must return handler error, NOT raw MCP error:
+
+🔴 16. `sqlite_transaction_begin({})` → handler error (or success if no required params)
+🔴 17. `sqlite_transaction_status({})` → handler error (or success if no required params)
+🔴 18. `sqlite_transaction_commit({})` → handler error (or success if no required params)
+🔴 19. `sqlite_transaction_rollback({})` → handler error (or success if no required params)
+🔴 20. `sqlite_transaction_execute({})` → handler error
+🔴 21. `sqlite_transaction_savepoint({})` → handler error
+🔴 22. `sqlite_transaction_release({})` → handler error
+🔴 23. `sqlite_transaction_rollback_to({})` → handler error
+
+---
+
+## Post-Test Procedures
+
+1. **Triage findings**: Create implementation plan if issues found
+2. **Scope of fixes**: Handler code, server-instructions, test database, this prompt
+3. **Validate**: Instruct the user to run the test suite (Vitest/Playwright), lint, and typecheck. Do NOT run them yourself.
+4. **Commit**: Stage and commit — do NOT push
+5. **Live re-test**: After server rebuild
+6. **Final summary**: After testing/re-testing
+
+---
+
+## Troubleshooting
+
+### Database is locked / file in use
+
+1. Check for Node.js processes: `Get-CimInstance Win32_Process -Filter "Name = 'node.exe'"`
+2. WAL/journal files are normal
+
+### Reset script fails
+
+1. Run with `-Verbose`: `.\reset-database.ps1 -Verbose`

package/test-server/test-tool-groups/test-vector-read.md

@@ -0,0 +1,130 @@
+# db-mcp Tool Group Testing: [vector-read]
+
+> [!IMPORTANT]
+> **Do not track progress in this file.** Track your test progress, coverage matrix, and findings in your internal task tracking system (artifact). However, you SHOULD edit this file to fix any factual errors, broken code, or incorrect assertions in the test prompts.
+> If there is nothing to fix, don't update UNRELEASED.md.
+> We're currently testing Native mode.
+
+## WASM Mode
+
+> When testing against a **WASM backend** (`sqlite-wasm` / sql.js): All 7 vector-read tools are fully WASM-compatible. No items to skip or adjust.
+
+**Step 1:** Confirm you read the server help content sourced from `C:\Users\chris\Desktop\db-mcp\src\constants\server-instructions\gotchas.md` using `view_file` (not grep or search) — to understand documented behaviors, edge cases, and response structures for this tool group.
+
+**Step 2:** Please conduct an exhaustive test of the **vector-read** tool group specified in the group-specific checklist below using live MCP server tool calls directly — not scripts/terminal.
+
+**Step 3:** The agent should update C:\Users\chris\Desktop\db-mcp\UNRELEASED.md with any/all changes/fixes.
+
+**Note** If temp tables are present from a previous test pass, it's because the database is locked. Ignore them.
+
+## Reporting Format
+
+- ❌ Fail: Tool errors or produces incorrect results (include error message)
+- ⚠️ Issue: Unexpected behavior or improvement opportunity
+- 📦 Payload: Unnecessarily large response that should be optimized.
+
+## Test Database Schema
+
+| Table             | Rows | Columns                                                                       | JSON Columns                                                                              |
+| ----------------- | ---- | ----------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------- |
+| test_products     | 16   | id, name, description, price, category, created_at                            | —                                                                                         |
+| test_orders       | 20   | id, product_id (FK), customer_name, quantity, total_price, order_date, status | —                                                                                         |
+| test_jsonb_docs   | 6    | id, doc, metadata, tags, created_at                                           | **doc**, **metadata** (nested), **tags** (array)                                          |
+| test_articles     | 8    | id, title, body, author, category, published_at                               | —                                                                                         |
+| test_users        | 9    | id, username, email, phone, bio, created_at                                   | —                                                                                         |
+| test_measurements | 200  | id, sensor_id, temperature, humidity, pressure, measured_at                   | —                                                                                         |
+| test_embeddings   | 20   | id, content, category, embedding                                              | **embedding** (8-dim float array); category values: database, fitness, food, tech, travel |
+| test_locations    | 15   | id, name, city, latitude, longitude, type                                     | —                                                                                         |
+| test_categories   | 17   | id, name, path, level                                                         | —                                                                                         |
+| test_events       | 100  | id, event_type, user_id (INT, 8 values), payload, event_date                  | **payload** (JSON)                                                                        |
+
+## Testing Requirements
+
+> [!CAUTION]
+> **Zero tolerance for raw MCP errors.** ANY response that is a raw MCP error (e.g., `-32602`, `isError: true`, no `success` field) is a **bug that must be reported and fixed**.
+
+1. Use existing `test_*` tables for read operations
+2. **Error path testing**: For **every** tool, test (a) domain error and (b) Zod validation error (`{}`). Both must return `{success: false, error: "..."}`.
+3. **Output schema testing**: For tools with `outputSchema`, confirm valid calls return structured JSON.
+
+## Structured Error Response Pattern
+
+```json
+{ "success": false, "error": "Human-readable error message" }
+```
+
+| Type                 | Source                                                             | What you see                                                | Verdict            |
+| -------------------- | ------------------------------------------------------------------ | ----------------------------------------------------------- | ------------------ |
+| **Handler error** ✅ | Handler catches error and returns `{success: false, error: "..."}` | Parseable JSON object with `success` and `error` fields     | Correct            |
+| **MCP error** ❌     | Uncaught throw propagates to MCP framework                         | Raw text error string, `isError: true` — no `success` field | Bug — report as ❌ |
+
+### Zod Validation Errors
+
+**Zod refinement leak pattern:** `.partial()` does NOT strip `.min(N)` / `.max(N)` refinements. **Fix:** Remove refinements from schema, validate inside handler.
+
+### Output Schema Validation Errors
+
+If valid inputs return raw MCP `-32602` mentioning "output schema", report as ❌.
+
+---
+
+## Group Focus: vector-read
+
+> **Instructions**: Execute every numbered checklist item with the exact inputs shown. Compare responses against the expected results. Report any deviation.
+
+> **Note:** Vector tools use pure JS computations (cosine, euclidean, dot product) with JSON-stored vectors — no native SQLite extension required. All 11 tools work identically in both WASM and Native modes.
+
+### vector-read Group Tools (7)
+
+1. sqlite_vector_search
+2. sqlite_vector_get
+3. sqlite_vector_count
+4. sqlite_vector_stats
+5. sqlite_vector_dimensions
+6. sqlite_vector_distance
+7. sqlite_vector_normalize
+
+**Test data:** `test_embeddings` (20 rows, 8-dim vectors, categories: tech, database, food, fitness, travel). Row 1: content="Machine learning fundamentals", embedding=[0.12, 0.45, -0.23, 0.78, 0.34, -0.56, 0.89, 0.01].
+
+**Checklist:**
+
+1. `sqlite_vector_count({table: "test_embeddings"})` → `{count: 20}`
+2. `sqlite_vector_dimensions({table: "test_embeddings", vectorColumn: "embedding"})` → `{dimensions: 8}`
+3. `sqlite_vector_get({table: "test_embeddings", idColumn: "id", vectorColumn: "embedding", id: 1})` → verify content="Machine learning fundamentals", category="tech", embedding has 8 dimensions
+4. `sqlite_vector_search({table: "test_embeddings", vectorColumn: "embedding", queryVector: [0.12, 0.45, -0.23, 0.78, 0.34, -0.56, 0.89, 0.01], metric: "cosine", limit: 3})` → top result should be row 1 (exact match, \_similarity ≈ 1)
+5. `sqlite_vector_search({table: "test_embeddings", vectorColumn: "embedding", queryVector: [0.12, 0.45, -0.23, 0.78, 0.34, -0.56, 0.89, 0.01], metric: "cosine", limit: 3, whereClause: "category = 'database'"})` → only database category results
+6. `sqlite_vector_distance({vector1: [1, 0, 0], vector2: [0, 1, 0], metric: "cosine"})` → distance ≈ 1.0 (orthogonal vectors)
+7. `sqlite_vector_distance({vector1: [3, 4], vector2: [0, 0], metric: "euclidean"})` → distance = 5.0
+8. `sqlite_vector_normalize({vector: [3, 4]})` → `{normalized: [0.6, 0.8], originalMagnitude: 5}`
+9. `sqlite_vector_stats({table: "test_embeddings", vectorColumn: "embedding"})` → verify min/max/avg magnitude
+
+**Code mode testing:**
+
+10. `sqlite_execute_code({code: "const result = await sqlite.vector.count({table: 'test_embeddings'}); return result;"})` → `{count: 20}`
+11. `sqlite_execute_code({code: "const result = await sqlite.vector.distance({vector1: [3, 4], vector2: [0, 0], metric: 'euclidean'}); return result;"})` → distance = 5.0
+
+**Error path testing:**
+
+🔴 12. `sqlite_vector_search({table: "nonexistent_table_xyz", vectorColumn: "embedding", queryVector: [1,2,3], metric: "cosine"})` → structured error
+🔴 13. `sqlite_vector_distance({vector1: [1, 2, 3], vector2: [1, 2], metric: "cosine"})` → error about dimension mismatch
+
+**Zod validation sweep** — call each tool with `{}` (empty params). Must return handler error, NOT raw MCP error:
+
+🔴 14. `sqlite_vector_search({})` → handler error
+🔴 15. `sqlite_vector_get({})` → handler error
+🔴 16. `sqlite_vector_count({})` → handler error
+🔴 17. `sqlite_vector_stats({})` → handler error
+🔴 18. `sqlite_vector_dimensions({})` → handler error
+🔴 19. `sqlite_vector_normalize({})` → handler error
+🔴 20. `sqlite_vector_distance({})` → handler error
+
+---
+
+## Post-Test Procedures
+
+1. **Triage findings**: Create implementation plan if issues found
+2. **Scope of fixes**: Handler code, server-instructions, test database, this prompt
+3. **Validate**: Instruct the user to run the test suite (Vitest/Playwright), lint, and typecheck. Do NOT run them yourself.
+4. **Commit**: Stage and commit — do NOT push
+5. **Live re-test**: After server rebuild
+6. **Final summary**: After testing/re-testing

package/test-server/test-tool-groups/test-vector-write.md

@@ -0,0 +1,115 @@
+# db-mcp Tool Group Testing: [vector-write]
+
+> [!IMPORTANT]
+> **Do not track progress in this file.** Track your test progress, coverage matrix, and findings in your internal task tracking system (artifact). However, you SHOULD edit this file to fix any factual errors, broken code, or incorrect assertions in the test prompts.
+> If there is nothing to fix, don't update UNRELEASED.md.
+> We're currently testing Native mode.
+
+## WASM Mode
+
+> When testing against a **WASM backend** (`sqlite-wasm` / sql.js): All 4 vector-write tools are fully WASM-compatible. No items to skip or adjust.
+
+**Step 1:** Confirm you read the server help content sourced from `C:\Users\chris\Desktop\db-mcp\src\constants\server-instructions\gotchas.md` using `view_file` (not grep or search) — to understand documented behaviors, edge cases, and response structures for this tool group.
+
+**Step 2:** Please conduct an exhaustive test of the **vector-write** tool group specified in the group-specific checklist below using live MCP server tool calls directly — not scripts/terminal.
+
+**Step 3:** The agent should update C:\Users\chris\Desktop\db-mcp\UNRELEASED.md with any/all changes/fixes.
+
+**Note** If temp tables are present from a previous test pass, it's because the database is locked. Ignore them.
+
+## Reporting Format
+
+- ❌ Fail: Tool errors or produces incorrect results (include error message)
+- ⚠️ Issue: Unexpected behavior or improvement opportunity
+- 📦 Payload: Unnecessarily large response that should be optimized.
+
+## Test Database Schema
+
+| Table             | Rows | Columns                                                                       | JSON Columns                                                                              |
+| ----------------- | ---- | ----------------------------------------------------------------------------- | ----------------------------------------------------------------------------------------- |
+| test_products     | 16   | id, name, description, price, category, created_at                            | —                                                                                         |
+| test_orders       | 20   | id, product_id (FK), customer_name, quantity, total_price, order_date, status | —                                                                                         |
+| test_jsonb_docs   | 6    | id, doc, metadata, tags, created_at                                           | **doc**, **metadata** (nested), **tags** (array)                                          |
+| test_articles     | 8    | id, title, body, author, category, published_at                               | —                                                                                         |
+| test_users        | 9    | id, username, email, phone, bio, created_at                                   | —                                                                                         |
+| test_measurements | 200  | id, sensor_id, temperature, humidity, pressure, measured_at                   | —                                                                                         |
+| test_embeddings   | 20   | id, content, category, embedding                                              | **embedding** (8-dim float array); category values: database, fitness, food, tech, travel |
+| test_locations    | 15   | id, name, city, latitude, longitude, type                                     | —                                                                                         |
+| test_categories   | 17   | id, name, path, level                                                         | —                                                                                         |
+| test_events       | 100  | id, event_type, user_id (INT, 8 values), payload, event_date                  | **payload** (JSON)                                                                        |
+
+## Testing Requirements
+
+> [!CAUTION]
+> **Zero tolerance for raw MCP errors.** ANY response that is a raw MCP error (e.g., `-32602`, `isError: true`, no `success` field) is a **bug that must be reported and fixed**.
+
+1. Use existing `test_*` tables for read operations
+2. Create temporary tables with `temp_*` prefix for write operations
+3. **Error path testing**: For **every** tool, test (a) domain error and (b) Zod validation error (`{}`). Both must return `{success: false, error: "..."}`.
+4. **Output schema testing**: For tools with `outputSchema`, confirm valid calls return structured JSON.
+
+## Structured Error Response Pattern
+
+```json
+{ "success": false, "error": "Human-readable error message" }
+```
+
+| Type                 | Source                                                             | What you see                                                | Verdict            |
+| -------------------- | ------------------------------------------------------------------ | ----------------------------------------------------------- | ------------------ |
+| **Handler error** ✅ | Handler catches error and returns `{success: false, error: "..."}` | Parseable JSON object with `success` and `error` fields     | Correct            |
+| **MCP error** ❌     | Uncaught throw propagates to MCP framework                         | Raw text error string, `isError: true` — no `success` field | Bug — report as ❌ |
+
+### Zod Validation Errors
+
+**Zod refinement leak pattern:** `.partial()` does NOT strip `.min(N)` / `.max(N)` refinements. **Fix:** Remove refinements from schema, validate inside handler.
+
+### Output Schema Validation Errors
+
+If valid inputs return raw MCP `-32602` mentioning "output schema", report as ❌.
+
+---
+
+## Group Focus: vector-write
+
+> **Instructions**: Execute every numbered checklist item with the exact inputs shown. Compare responses against the expected results. Report any deviation.
+
+### vector-write Group Tools (4)
+
+1. sqlite_vector_create_table
+2. sqlite_vector_store
+3. sqlite_vector_batch_store
+4. sqlite_vector_delete
+
+**Checklist:**
+
+**Write operations (use temp tables):**
+
+1. `sqlite_vector_create_table({tableName: "temp_vector_test", dimensions: 8, additionalColumns: [{name: "content", type: "TEXT"}, {name: "category", type: "TEXT"}]})` → success
+2. `sqlite_vector_store({table: "temp_vector_test", idColumn: "id", vectorColumn: "vector", id: 1, vector: [0.1, 0.2, 0.3, 0.4, 0.5, 0.6, 0.7, 0.8]})` → success
+3. `sqlite_vector_batch_store({table: "temp_vector_test", idColumn: "id", vectorColumn: "vector", items: [{id: 2, vector: [0.11, 0.22, 0.33, 0.44, 0.55, 0.66, 0.77, 0.88]}, {id: 3, vector: [0.9, 0.8, 0.7, 0.6, 0.5, 0.4, 0.3, 0.2]}]})` → `{stored: 2}`
+4. Verify rows stored: `sqlite_vector_count({table: "temp_vector_test"})` → `{count: 3}` (use the direct tool or code mode)
+5. `sqlite_vector_delete({table: "temp_vector_test", idColumn: "id", ids: [1]})` → success
+6. Verify deletion: `sqlite_vector_count({table: "temp_vector_test"})` → `{count: 2}`
+7. Cleanup: `sqlite_drop_table({table: "temp_vector_test"})` (using core tool)
+
+**Error path testing:**
+
+🔴 8. `sqlite_vector_store({table: "nonexistent_table_xyz", idColumn: "id", vectorColumn: "vec", id: 1, vector: [1,2,3]})` → structured error
+
+**Zod validation sweep** — call each tool with `{}` (empty params). Must return handler error, NOT raw MCP error:
+
+🔴 9. `sqlite_vector_create_table({})` → handler error
+🔴 10. `sqlite_vector_store({})` → handler error
+🔴 11. `sqlite_vector_batch_store({})` → handler error
+🔴 12. `sqlite_vector_delete({})` → handler error
+
+---
+
+## Post-Test Procedures
+
+1. **Triage findings**: Create implementation plan if issues found
+2. **Scope of fixes**: Handler code, server-instructions, test database, this prompt
+3. **Validate**: Instruct the user to run the test suite (Vitest/Playwright), lint, and typecheck. Do NOT run them yourself.
+4. **Commit**: Stage and commit — do NOT push
+5. **Live re-test**: After server rebuild
+6. **Final summary**: After testing/re-testing