memory-journal-mcp 6.1.2 → 6.2.1

package/test-server/test-tools.md

@@ -1,429 +0,0 @@
-# Test memory-journal-mcp — Core & Mutations
-
-Exhaustively test the memory-journal-mcp server's core functionality using the phased plan below. **Please make sure to use the correct resource names/urls as documented below.**
-
-**Scope:** 24 core tools (entry CRUD, search, analytics, relationships, admin, backup), seed data — this file covers happy paths, core error paths, and feature verification (Phases 0-6). GitHub tools (16 tools), all 27 resources, and prompt handlers are tested in `test-tools2.md`. Team tools (20 tools) are tested in `test-tools-team.md`.
-
-**Seed data:** S1–S12 are general-purpose seeds. S13–S14 are personal journal entries for `project_number: 5` (needed alongside S7 to reach the `min_entries: 3` threshold for `get_cross_project_insights`). S15–S17 are team DB entries for `project_number: 5` (needed for `team_get_cross_project_insights`).
-
-**Constraints:**
-
-- Confirm MCP server instructions were auto-received before starting.
-- Use the MCP server directly for all tests — not the terminal or scripts.
-- Use https://github.com/users/neverinfamous/projects/5 for project/Kanban testing.
-
-**Workflow after testing:**
-
-1. Create a plan to implement any fixes and/or improvement opportunities, including changes to `server-instructions.md`/`server-instructions.ts` or this file (`test-server/test-tools.md`).
-2. Use `code-map.md` as a source of truth and ensure fixes comply with `C:\Users\chris\Desktop\adamic\skills\mcp-builder`.
-3. After implementation, update `UNRELEASED.md` and commit without pushing. Then, stop so the user can verify with `npm run lint && npm run typecheck`, `npm run test`, and `npm run test:e2e`.
-4. After user completes verification, re-test fixes with direct MCP calls.
-5. Provide a very brief final summary.
-
-> [!IMPORTANT]
-> **Test Session Prerequisites**
-
-1. The server instructions are auto-injected by the MCP protocol. Confirm receipt (no need to read `memory://instructions` separately).
-2. Confirm `memory://briefing` was auto-received and **present the `userMessage` to the user as a formatted bullet list of key facts as the server instructions required:**. Detailed briefing testing is below in Phase 1.2.
-
----
-
-## Phase 0: Seed Test Data
-
-> [!IMPORTANT]
-> Create these entries **before Phase 2** to ensure FTS5, filter, semantic search, and relationship tests have matching data. Each entry is mapped to the test cases it enables.
->
-> **Do NOT skip this phase** — without it, many Phase 3 search tests will return empty results and won't validate the underlying feature.
-
-### 0.1 FTS5 Content Entries
-
-These entries ensure every FTS5 query pattern in Phase 3.1 returns actual results.
-
-| #   | Tool           | Params                                                                                                                                                                                              | Enables Tests                                                      |
-| --- | -------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------ |
-| S1  | `create_entry` | `content: "Redesigned the authentication architecture for the OAuth 2.1 module"`, `entry_type: "technical_note"`, `tags: ["architecture", "auth"]`                                                  | `architecture`, `auth*`, `"authentication architecture"` phrase    |
-| S2  | `create_entry` | `content: "Improved error handling in the database adapter layer with typed error classes"`, `entry_type: "bug_fix"`, `significance_type: "lesson_learned"`, `tags: ["error-handling", "database"]` | `"error handling"` phrase, `significance_type` filter              |
-| S3  | `create_entry` | `content: "Deploy new release candidate to the CDN edge network"`, `entry_type: "feature_implementation"`, `tags: ["deploy", "release"]`, `is_personal: false`                                      | `deploy NOT staging`, `deploy OR release`                          |
-| S4  | `create_entry` | `content: "Released v5.0 with breaking API changes and migration guide"`, `entry_type: "milestone"`, `significance_type: "breakthrough"`, `tags: ["release"]`                                       | `deploy OR release` (via "release"), semantic search for "release" |
-| S5  | `create_entry` | `content: "Deploy to staging environment failed — rollback initiated"`, `entry_type: "bug_fix"`, `tags: ["deploy", "staging"]`                                                                      | `deploy NOT staging` (negative match — verifies NOT exclusion)     |
-| S6  | `create_entry` | `content: "The test's scope was expanded to cover 100% of edge cases"`, `entry_type: "planning"`, `tags: ["testing"]`                                                                               | `test's` LIKE fallback, `100%` LIKE fallback                       |
-
-### 0.2 Filter & GitHub-Linked Entries
-
-These entries ensure filter tests (`issue_number`, `pr_status`, `workflow_run_id`, `project_number`) return results.
-
-| #   | Tool           | Params                                                                                                                                                                                                                                     | Enables Tests                                                                            |
-| --- | -------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | ---------------------------------------------------------------------------------------- |
-| S7  | `create_entry` | `content: "Investigated performance regression in issue #44 — root cause was N+1 queries"`, `entry_type: "research"`, `issue_number: 44`, `project_number: 5`, `tags: ["performance", "investigation"]`                                    | `issue_number: 44` filter, `project_number: 5` filter, semantic search for "performance" |
-| S8  | `create_entry` | `content: "Code review feedback from PR #67 merged — refactored authentication middleware"`, `entry_type: "code_review"`, `pr_number: 67`, `pr_status: "merged"`, `tags: ["code-review", "auth"]`                                          | `pr_status: "merged"` filter, `pr_number` filter, `auth*` prefix                         |
-| S9  | `create_entry` | `content: "CI workflow run completed — all 910 tests passing across 3 test suites"`, `entry_type: "technical_note"`, `workflow_run_id: 12345`, `workflow_name: "lint-and-test"`, `workflow_status: "completed"`, `tags: ["ci", "testing"]` | `workflow_run_id` filter, workflow field persistence                                     |
-| S10 | `create_entry` | `content: "Personal reflection on improving development velocity and reducing technical debt"`, `entry_type: "personal_reflection"`, `is_personal: true`, `tags: ["personal", "velocity"]`                                                 | `is_personal: true` filter, semantic search for "improving performance"                  |
-
-### 0.3 Team & Cross-DB Entries
-
-These entries ensure cross-DB search merging (`source: 'personal' | 'team'`) returns results from both databases.
-
-| #   | Tool                | Params                                                                                                                                                                                     | Enables Tests                                                                    |
-| --- | ------------------- | ------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------ | -------------------------------------------------------------------------------- |
-| S11 | `create_entry`      | `content: "Architecture decision: adopted event-driven design for webhook processing"`, `entry_type: "project_decision"`, `share_with_team: true`, `tags: ["architecture", "team-shared"]` | Cross-DB `search_entries` with `source` marker, team search, `architecture` FTS5 |
-| S12 | `team_create_entry` | `content: "Team standup: discussed authorization flow improvements and deploy pipeline"`, `entry_type: "standup"`, `tags: ["standup", "auth", "deploy"]`                                   | Team-only search, cross-DB date range, `auth*` and `deploy` in team DB           |
-
-### 0.5 Cross-Project Insights Seed
-
-> [!IMPORTANT]
-> `get_cross_project_insights` and `team_get_cross_project_insights` require a **minimum of 3 entries** sharing the same `project_number` before they appear in results (default `min_entries: 3`). S7 already has `project_number: 5` — add S13–S14 to reach the threshold in the personal journal. S15–S17 seed the team DB with project-linked entries.
-
-**Personal journal — complete project #5 to 3 entries:**
-
-| #   | Tool           | Params                                                                                                                                                                                               | Enables Tests                                                                     |
-| --- | -------------- | ---------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------- |
-| S13 | `create_entry` | `content: "Completed sprint planning for project #5: scoped auth and deploy milestones"`, `entry_type: "planning"`, `project_number: 5`, `tags: ["planning", "sprint"]`                              | `project_number: 5` filter, `get_cross_project_insights` non-zero `project_count` |
-| S14 | `create_entry` | `content: "Project #5 retrospective: identified bottlenecks in deployment pipeline and action items"`, `entry_type: "personal_reflection"`, `project_number: 5`, `tags: ["retrospective", "deploy"]` | `get_cross_project_insights` — 3rd entry to meet `min_entries: 3` for project 5   |
-
-**Team DB — seed 3 project-linked entries:**
-
-| #   | Tool                | Params                                                                                                                                                                                              | Enables Tests                                                                         |
-| --- | ------------------- | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------- |
-| S15 | `team_create_entry` | `content: "Team kickoff for project #5: aligned on goals and delivery timeline"`, `entry_type: "standup"`, `project_number: 5`, `tags: ["kickoff", "project"]`                                      | `team_get_cross_project_insights` non-zero `project_count`                            |
-| S16 | `team_create_entry` | `content: "Project #5 mid-sprint check-in: auth module ahead of schedule, deploy pipeline at risk"`, `entry_type: "standup"`, `project_number: 5`, `tags: ["standup", "project"]`                   | `team_get_cross_project_insights` — 2nd team entry for project 5                      |
-| S17 | `team_create_entry` | `content: "Project #5 release review: all acceptance criteria met, feature flags enabled for rollout"`, `entry_type: "standup"`, `project_number: 5`, `tags: ["release", "project", "team-shared"]` | `team_get_cross_project_insights` — 3rd team entry to meet `min_entries: 3` threshold |
-
-### 0.4 Post-Seed Verification
-
-After creating all 17 entries, verify the seed data is searchable:
-
-| Check                       | Command                                           | Expected                                                                                                            |
-| --------------------------- | ------------------------------------------------- | ------------------------------------------------------------------------------------------------------------------- |
-| FTS5 indexed                | `search_entries(query: "architecture")`           | ≥ 1 result (S1 or S11 depending on BM25 rank); use phrase `"authentication architecture"` to ensure S1 specifically |
-| Filters work                | `search_entries(issue_number: 44)`                | ≥ 1 result (S7)                                                                                                     |
-| Cross-DB merged             | `search_entries(query: "architecture")`           | At least 1 result includes `source: 'team'` (S11); use `auth*` for cross-DB results spanning both DBs               |
-| Rebuild vector index        | `rebuild_vector_index`                            | `entriesIndexed` > 0                                                                                                |
-| Semantic search             | `semantic_search(query: "improving performance")` | ≥ 1 result (S7, S10 should be semantically similar)                                                                 |
-| Cross-project insights      | `get_cross_project_insights({})`                  | `project_count ≥ 1`, project 5 appears with `entry_count ≥ 3`                                                       |
-| Team cross-project insights | `team_get_cross_project_insights({})`             | `project_count ≥ 1`, project 5 appears with `entry_count ≥ 3`                                                       |
-
----
-
-## Phase 1: Core Infrastructure
-
-### 1.1 Server Health & Connectivity
-
-| Test               | Command/Action         | Expected Result                                                                                 |
-| ------------------ | ---------------------- | ----------------------------------------------------------------------------------------------- |
-| Basic connectivity | `test_simple`          | Returns echo message                                                                            |
-| Health resource    | Read `memory://health` | Shows DB stats, tool filter status, vector index health                                         |
-| Health team block  | Read `memory://health` | Includes `teamDatabase` block with `configured`, `entryCount`, `path` (requires `TEAM_DB_PATH`) |
-| Scheduler inactive | Read `memory://health` | `scheduler.active: false`, `scheduler.jobs: []` (scheduler is HTTP-only, inactive in stdio)     |
-| Statistics         | `get_statistics`       | Returns `structuredContent` with entry counts, types, periods                                   |
-| Enhanced analytics | `get_statistics`       | Returns `decisionDensity`, `relationshipComplexity`, `activityTrend`, `causalMetrics`           |
-
-### 1.2 Briefing Resource
-
-| Test                             | Command/Action                   | Expected Result                                                                                        |
-| -------------------------------- | -------------------------------- | ------------------------------------------------------------------------------------------------------ |
-| Read briefing                    | Read `memory://briefing`         | Returns JSON with `userMessage`, `templateResources`, `journal`, `github`                              |
-| Verify `lastModified` annotation | Check resource metadata          | ISO 8601 timestamp (client-dependent — AntiGravity doesn't expose MCP annotations)                     |
-| Confirm `userMessage`            | Inspect briefing.userMessage     | Formatted table with project/branch/CI/journal stats                                                   |
-| Milestone progress row           | Inspect briefing.userMessage     | Table includes milestone progress row (e.g., "🚩 Milestones: X open")                                  |
-| Team DB row                      | Inspect briefing.userMessage     | Table includes "Team DB" row with team entry count (requires `TEAM_DB_PATH`)                           |
-| Template URIs                    | Check `templateResources` array  | 7 template URIs listed (includes `memory://milestones/{number}`)                                       |
-| Workflow summary                 | Inspect `github.workflowSummary` | Present when `BRIEFING_WORKFLOW_STATUS=true` — has `passing`, `failing`, `pending`, `cancelled` counts |
-| Workflow named runs              | Inspect `workflowSummary.runs`   | Array of `{name, conclusion}` when `BRIEFING_WORKFLOW_COUNT > 0`; CI row shows icons (✅/❌)           |
-| Rules metadata                   | Inspect `rulesFile` field        | Present when `RULES_FILE_PATH` set — has `name`, `sizeKB`, `lastModified`                              |
-| Skills metadata                  | Inspect `skillsDir` field        | Present when `SKILLS_DIR_PATH` set — has `count`, `names` array                                        |
-| Enhanced CI row                  | Inspect briefing.userMessage     | CI row shows breakdown or named runs (not just single-word status) when workflow env vars are set      |
-
-### 1.3 Protocol Validation — Run via Scripts - DO NOT SKIP!
-
-> [!IMPORTANT]
-> These tests require **separate server starts** — they cannot be run via MCP tool calls. Run the scripts below in a terminal. See `test-server/README.md` for full details and script locations.
-
-```powershell
-# Test A — Instruction levels (essential < standard < full)
-node test-server/test-instruction-levels.mjs
-
-# Test B — Tool annotations (61 tools, 45 openWorldHint=false, 16 openWorldHint=true, 0 missing)
-node test-server/test-tool-annotations.mjs
-```
-
-| Check              | Expected                                                             |
-| ------------------ | -------------------------------------------------------------------- |
-| Instruction levels | essential (~1.5K) < standard (~1.7K) < full (~2.7K tokens)           |
-| Tool annotations   | 61 tools, all with `annotations`, 45 `false` + 16 `true` = 0 missing |
-
-### 1.4 GitHub Status Resource
-
-| Test              | Command/Action                | Expected Result                                                  |
-| ----------------- | ----------------------------- | ---------------------------------------------------------------- |
-| Read status       | Read `memory://github/status` | Compact JSON with repo, branch, CI, issues, PRs, Kanban summary  |
-| CI status mapping | Verify CI status value        | Shows `passing`, `failing`, `pending`, `cancelled`, or `unknown` |
-| Milestone data    | Inspect status data           | Includes milestones summary (open count, completion percentages) |
-
----
-
-## Phase 2: Entry CRUD Operations
-
-### 2.1 Create Entry
-
-| Test                 | Command/Action                                                                                              | Expected Result                                                                       |
-| -------------------- | ----------------------------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------- |
-| Minimal create       | `create_entry_minimal(content: "Test entry")`                                                               | Returns entry ID                                                                      |
-| Full create          | `create_entry(content: "...", entry_type: "planning", tags: ["test"])`                                      | Entry created with metadata                                                           |
-| With GitHub linking  | `create_entry(..., issue_number: <N>)`                                                                      | Entry links to issue                                                                  |
-| issueUrl auto-pop    | `create_entry(content: "...", issue_number: <N>)` — omit issueUrl                                           | `issueUrl` auto-populated from cached repo info (requires prior `get_github_context`) |
-| Invalid entry_type   | `create_entry(content: "test", entry_type: "invalid")`                                                      | Structured error: `{ success: false, error: "..." }` listing valid enum values        |
-| Invalid significance | `create_entry(content: "test", significance_type: "invalid")`                                               | Structured error: `{ success: false, error: "..." }` listing valid enum values        |
-| With PR fields       | `create_entry(content: "PR test", pr_number: 67, pr_status: "merged")`                                      | Entry created with `prNumber`, `prStatus` fields persisted                            |
-| With workflow fields | `create_entry(content: "CI test", workflow_run_id: 123, workflow_name: "CI", workflow_status: "completed")` | Entry created with all workflow fields persisted                                      |
-| With project_owner   | `create_entry(content: "...", project_number: 5, project_owner: "neverinfamous")`                           | Entry created with `projectOwner` field                                               |
-| auto_context off     | `create_entry(content: "No context", auto_context: false)`                                                  | Entry created without auto-generated context                                          |
-| share_with_team      | `create_entry(content: "Shared entry", share_with_team: true)`                                              | Entry in personal DB + team DB; response has `sharedWithTeam: true`, `author`         |
-
-### 2.2 Read & Update
-
-| Test                      | Command/Action                                                 | Expected Result                                                          |
-| ------------------------- | -------------------------------------------------------------- | ------------------------------------------------------------------------ |
-| Get by ID                 | `get_entry_by_id(entry_id: <N>)`                               | Returns `structuredContent` with relationships                           |
-| Importance score          | `get_entry_by_id(entry_id: <N>)`                               | Response includes `importance` field (0.0-1.0) and `importanceBreakdown` |
-| No relationships          | `get_entry_by_id(entry_id: <N>, include_relationships: false)` | Response omits `relationships` array (or returns empty)                  |
-| Update tags               | `update_entry(entry_id: <N>, tags: ["updated"])`               | Tags changed                                                             |
-| Update content            | `update_entry(entry_id: <N>, content: "Updated content")`      | Content changed; verify via `get_entry_by_id`                            |
-| Update entry_type         | `update_entry(entry_id: <N>, entry_type: "technical_note")`    | Entry type changed                                                       |
-| Update is_personal        | `update_entry(entry_id: <N>, is_personal: false)`              | `isPersonal` toggled                                                     |
-| Update nonexistent        | `update_entry(entry_id: 999999, tags: ["x"])`                  | Returns `{ success: false, error: "Entry 999999 not found" }`            |
-| Update invalid type       | `update_entry(entry_id: <N>, entry_type: "invalid")`           | Structured error listing valid enum values                               |
-| Get recent                | `get_recent_entries(limit: 5)`                                 | Returns `structuredContent` array                                        |
-| Get recent (personal)     | `get_recent_entries(limit: 5, is_personal: true)`              | Only personal entries returned                                           |
-| Get recent (not personal) | `get_recent_entries(limit: 5, is_personal: false)`             | Only non-personal entries returned                                       |
-
-### 2.3 Delete (Test Last!)
-
-| Test                     | Command/Action                                        | Expected Result                                           |
-| ------------------------ | ----------------------------------------------------- | --------------------------------------------------------- |
-| Soft delete              | `delete_entry(entry_id: <test_id>, permanent: false)` | Entry hidden from search                                  |
-| Permanent delete         | `delete_entry(entry_id: <test_id>, permanent: true)`  | Entry removed                                             |
-| Delete nonexistent entry | `delete_entry(entry_id: 999999, permanent: false)`    | Returns `success: false, error: "Entry 999999 not found"` |
-
----
-
-## Phase 3: Search Tools
-
-### 3.1 Text Search
-
-| Test                  | Command/Action                                                                                   | Expected Result                                                                                                                                                                               |
-| --------------------- | ------------------------------------------------------------------------------------------------ | --------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------------- |
-| FTS5 search           | `search_entries(query: "architecture")`                                                          | ≥ 1 result — S1 and S11 both indexed; BM25 may rank team entry first; use `search_entries(query: "authentication architecture")` to target S1 specifically                                    |
-| FTS5 phrase           | `search_entries(query: "\"error handling\"")`                                                    | ≥ 1 result (S2) — exact phrase match only; the query param value must contain the literal quotes as part of the string: `"error handling"` (not JSON escape sequences)                        |
-| FTS5 prefix           | `search_entries(query: "auth*")`                                                                 | ≥ 2 results (S1, S8) — matches "authentication", "authorization", etc.                                                                                                                        |
-| FTS5 boolean NOT      | `search_entries(query: "deploy NOT staging")`                                                    | Returns S3, S11 but NOT S5 (S5 contains "staging")                                                                                                                                            |
-| FTS5 boolean OR       | `search_entries(query: "deploy OR release")`                                                     | ≥ 2 results (S3, S4, S5 expected)                                                                                                                                                             |
-| FTS5 fallback         | `search_entries(query: "test's")`                                                                | ≥ 1 result (S6) — LIKE fallback, single quotes are FTS5-unsafe                                                                                                                                |
-| FTS5 special chars    | `search_entries(query: "100%")`                                                                  | ≥ 1 result (S6) — LIKE fallback, `%` is FTS5-unsafe                                                                                                                                           |
-| Date range            | `search_by_date_range(start_date: "2026-01-01", end_date: "2026-01-31")`                         | Returns `structuredContent` array                                                                                                                                                             |
-| Cross-DB search       | `search_entries(query: "test")`                                                                  | Results include `source: 'personal' \| 'team'` marker on each entry                                                                                                                           |
-| Cross-DB date         | `search_by_date_range(start_date: "2026-01-01", end_date: "2026-12-31")`                         | Results include `source` marker merging personal + team entries                                                                                                                               |
-| Invalid date fmt      | `search_by_date_range(start_date: "Jan 1", end_date: "Jan 31")`                                  | Structured error: `{ success: false, error: "..." }` with YYYY-MM-DD hint                                                                                                                     |
-| Filter by issue       | `search_entries(issue_number: 44)`                                                               | Returns entries linked to issue #44                                                                                                                                                           |
-| Filter by PR status   | `search_entries(pr_status: "merged")`                                                            | Returns entries with `prStatus: "merged"`                                                                                                                                                     |
-| Filter by workflow    | `search_entries(workflow_run_id: <N>)`                                                           | Returns entries linked to workflow run                                                                                                                                                        |
-| Filter by project     | `search_entries(project_number: 5)`                                                              | Returns entries linked to project #5                                                                                                                                                          |
-| Filter by is_personal | `search_entries(query: "test", is_personal: true)`                                               | Only personal entries returned                                                                                                                                                                |
-| Date range + type     | `search_by_date_range(start_date: "2026-01-01", end_date: "2026-12-31", entry_type: "planning")` | Only "planning" entries in date range                                                                                                                                                         |
-| Date range + tags     | `search_by_date_range(start_date: "2026-01-01", end_date: "2026-12-31", tags: ["test"])`         | Only entries with "test" tag in date range                                                                                                                                                    |
-| Date range + personal | `search_by_date_range(start_date: "2026-01-01", end_date: "2026-12-31", is_personal: true)`      | Only personal entries in date range                                                                                                                                                           |
-| Date range + project  | `search_by_date_range(start_date: "2026-01-01", end_date: "2026-12-31", project_number: 5)`      | Only project #5 entries in date range                                                                                                                                                         |
-| Inverted date range   | `search_by_date_range(start_date: "2026-12-31", end_date: "2026-01-01")`                         | Returns `{ success: false, error: "Invalid date range: start_date (...) is after end_date (...)", code: "VALIDATION_ERROR", suggestion: "Ensure start_date is before or equal to end_date" }` |
-
-> [!NOTE]
-> **Cross-DB Search Behavior:** When a team DB is present, per-DB queries fetch `limit × 2` (capped at 500) to prevent BM25 ranking in one DB from silently dropping entries before the cross-DB merge. The user's requested `limit` is applied after merging.
->
-> **Code Mode API Group Structure:** When testing `mj_execute_code`, methods are bound to specific groups. Key mapping: `listTags` → `mj.core`, `mergeTags` → `mj.admin`, `getStatistics` → `mj.analytics`. Use `mj.help()` or `mj.<group>.help()` to discover available methods per group.
-
-### 3.2 Semantic Search
-
-| Test                   | Command/Action                                                     | Expected Result                                                                     |
-| ---------------------- | ------------------------------------------------------------------ | ----------------------------------------------------------------------------------- |
-| Vector index stats     | `get_vector_index_stats`                                           | Shows `itemCount`, `modelName`, `dimensions`                                        |
-| Rebuild index          | `rebuild_vector_index`                                             | `entriesIndexed` > 0 (indexes seed entries)                                         |
-| Semantic query         | `semantic_search(query: "improving performance")`                  | ≥ 1 result — S7, S10 semantically similar                                           |
-| Custom threshold       | `semantic_search(query: "performance", similarity_threshold: 0.5)` | Fewer results than default threshold (0.25)                                         |
-| Personal filter        | `semantic_search(query: "test", is_personal: true)`                | Only personal entries in results                                                    |
-| Hint disabled          | `semantic_search(query: "xyznonexistent", hint_on_empty: false)`   | Noise results with quality gate `hint` still shown (only advisory hints suppressed) |
-| Hint enabled (default) | `semantic_search(query: "xyznonexistent")`                         | Noise results with quality gate `hint` (all hints shown)                            |
-
-### 3.3 Analytics & Index Management
-
-| Test                     | Command/Action                                                                 | Expected Result                                                                          |
-| ------------------------ | ------------------------------------------------------------------------------ | ---------------------------------------------------------------------------------------- |
-| Stats group by month     | `get_statistics(group_by: "month")`                                            | Periods grouped by month                                                                 |
-| Stats group by day       | `get_statistics(group_by: "day")`                                              | Periods grouped by day                                                                   |
-| Stats with dates         | `get_statistics(start_date: "2026-01-01", end_date: "2026-03-01")`             | Returns `dateRange` in response; results filtered to date range                          |
-| Stats project breakdown  | `get_statistics(project_breakdown: true)`                                      | Returns `projectBreakdown` array with per-project stats                                  |
-| Cross-project insights   | `get_cross_project_insights`                                                   | `project_count ≥ 1`, `projects` array with project 5 (entry_count ≥ 3, top_tags present) |
-| Insights with dates      | `get_cross_project_insights(start_date: "2026-01-01", end_date: "2026-03-01")` | Date-filtered project insights — project 5 visible if S7/S13/S14 within range            |
-| Insights min_entries     | `get_cross_project_insights(min_entries: 1)`                                   | Same or more projects than default (min_entries: 3)                                      |
-| Add to vector index      | `add_to_vector_index(entry_id: <existing_id>)`                                 | `success: true`, `entryId` in response                                                   |
-| Add nonexistent to index | `add_to_vector_index(entry_id: 999999)`                                        | Returns `{ success: false, error: "..." }`                                               |
-
----
-
-## Phase 4: Relationships & Visualization
-
-### 4.1 Basic Relationships
-
-| Test                    | Command/Action                                                                                                            | Expected Result                                                                                           |
-| ----------------------- | ------------------------------------------------------------------------------------------------------------------------- | --------------------------------------------------------------------------------------------------------- |
-| Link entries            | `link_entries(from_entry_id: <A>, to_entry_id: <B>, relationship_type: "references")`                                     | Relationship created                                                                                      |
-| Duplicate link          | Call `link_entries` again with same params                                                                                | Returns `duplicate: true`, `message`, existing relationship                                               |
-| Link nonexistent source | `link_entries(from_entry_id: 999999, to_entry_id: <B>, ...)`                                                              | Returns `success: false`, message: `"One or both entries not found (from: 999999, to: <B>)"`              |
-| Link nonexistent target | `link_entries(from_entry_id: <A>, to_entry_id: 999999, ...)`                                                              | Returns `success: false`, message: `"One or both entries not found (from: <A>, to: 999999)"`              |
-| Visualize               | `visualize_relationships(entry_id: <A>)`                                                                                  | JSON object with `mermaid` string field containing diagram, `entry_count`, `relationship_count`, `legend` |
-| Link with description   | `link_entries(from_entry_id: <A>, to_entry_id: <C>, relationship_type: "implements", description: "Implements the plan")` | Relationship created with `description` field                                                             |
-| Reverse duplicate       | `link_entries(from_entry_id: <B>, to_entry_id: <A>, relationship_type: "references")`                                     | Succeeds — only same-direction duplicates are checked (confirmed)                                         |
-| Visualize nonexistent   | `visualize_relationships(entry_id: 999999)`                                                                               | Returns `message: "Entry 999999 not found"`                                                               |
-| Visualize by tags       | `visualize_relationships(tags: ["test"])`                                                                                 | Diagram scoped to entries with "test" tag                                                                 |
-| Visualize depth 3       | `visualize_relationships(entry_id: <A>, depth: 3)`                                                                        | Deeper traversal than default `depth: 2`                                                                  |
-| Visualize custom limit  | `visualize_relationships(entry_id: <A>, limit: 5)`                                                                        | Diagram limited to 5 entries                                                                              |
-| Graph resource          | Read `memory://graph/recent`                                                                                              | Raw Mermaid text (`text/plain` MIME), arrows harmonized with `visualize_relationships`                    |
-
-### 4.2 Causal Relationship Types
-
-| Test              | Command/Action                                                                        | Expected Result                                                                       |
-| ----------------- | ------------------------------------------------------------------------------------- | ------------------------------------------------------------------------------------- |
-| blocked_by type   | `link_entries(from_entry_id: <A>, to_entry_id: <B>, relationship_type: "blocked_by")` | Relationship created with `blocked_by` type                                           |
-| resolved type     | `link_entries(from_entry_id: <A>, to_entry_id: <B>, relationship_type: "resolved")`   | Relationship created with `resolved` type                                             |
-| caused type       | `link_entries(from_entry_id: <A>, to_entry_id: <B>, relationship_type: "caused")`     | Relationship created with `caused` type                                               |
-| Causal viz arrows | `visualize_relationships(entry_id: <A>)`                                              | Mermaid shows `--x` (blocked_by), `==>` (resolved), `-.->` (caused) arrows            |
-| Graph harmonized  | Read `memory://graph/recent`                                                          | Raw Mermaid `text/plain`: `--x`, `==>`, `-.->` for causal types, `-->` for references |
-
----
-
-## Phase 5: Admin & Backup Tools
-
-### 5.1 Tags
-
-| Test              | Command/Action                                                | Expected Result                                                             |
-| ----------------- | ------------------------------------------------------------- | --------------------------------------------------------------------------- |
-| List tags         | `list_tags`                                                   | Returns all tags with counts                                                |
-| Create source tag | `create_entry(content: "Test tag merge", tags: ["test-old"])` | Creates "test-old" tag (pre-req)                                            |
-| Merge tags        | `merge_tags(source_tag: "test-old", target_tag: "test-new")`  | Merges source into target, deletes source                                   |
-| Verify merge      | `list_tags` + `search_entries(query: "Test tag merge")`       | "test-old" gone, "test-new" exists, entry now has "test-new" tag            |
-| Merge same tag    | `merge_tags(source_tag: "test-new", target_tag: "test-new")`  | Structured error: `{ success: false, error: "..." }` (source equals target) |
-| Merge nonexistent | `merge_tags(source_tag: "nonexistent-xyz", target_tag: "x")`  | Structured error: `{ success: false, error: "Source tag not found: ..." }`  |
-
-> [!NOTE]
-> If `restore_backup` is tested after `merge_tags`, the restored backup will revert the merge. This is expected behavior. Verify merge worked immediately after calling `merge_tags`, before any backup restoration.
-
-### 5.2 Export
-
-| Test                    | Command/Action                                                                     | Expected Result                          |
-| ----------------------- | ---------------------------------------------------------------------------------- | ---------------------------------------- |
-| Export JSON             | `export_entries(format: "json", limit: 5)`                                         | JSON export with `entries` array         |
-| Export markdown         | `export_entries(format: "markdown", limit: 5)`                                     | Markdown export with `content` string    |
-| Export with tags        | `export_entries(format: "json", tags: ["test"], limit: 10)`                        | Only entries with matching tags returned |
-| Export with dates       | `export_entries(format: "json", start_date: "2026-01-01", end_date: "2026-03-01")` | Only entries within date range returned  |
-| Export with entry_types | `export_entries(format: "json", entry_types: ["planning"], limit: 10)`             | Only entries of specified type returned  |
-
-### 5.3 Backup & Restore
-
-| Test                  | Command/Action                                              | Expected Result                                                                  |
-| --------------------- | ----------------------------------------------------------- | -------------------------------------------------------------------------------- |
-| Create backup         | `backup_journal(name: "test-backup")`                       | Backup file created with `success`, `filename`, `path`, `sizeBytes`              |
-| Auto-named backup     | `backup_journal`                                            | Backup created with auto-generated timestamped name                              |
-| List backups          | `list_backups`                                              | Shows backup files with metadata including `path` field                          |
-| Cleanup backups       | `cleanup_backups(keep_count: 5)`                            | Deletes old backups, keeps N most recent                                         |
-| Backup path traversal | `backup_journal(name: "../../etc/passwd")`                  | Structured error: `{ success: false, error: "..." }` with path traversal message |
-| Restore backup        | `restore_backup(filename: "test-backup.db", confirm: true)` | Restores + `revertedChanges` field with details                                  |
-| Restore nonexistent   | `restore_backup(filename: "nonexistent.db", confirm: true)` | Structured error: `{ success: false, error: "Backup file not found: ..." }`      |
-
----
-
-## Phase 6: Automated Scheduler — Run via Script [DO NOT SKIP!]
-
-> [!IMPORTANT]
-> The scheduler only activates in HTTP/SSE transport mode. Run the script below — it handles session init, health reads, and wait/verify automatically. See `test-server/README.md` for full details.
-
-```powershell
-# Terminal 1: Start HTTP server with short scheduler intervals
-npm run build
-node dist/cli.js --transport http --port 3099 --backup-interval 1 --keep-backups 3 --vacuum-interval 2 --rebuild-index-interval 2
-
-# Terminal 2: Run scheduler test (waits 130s for jobs to fire)
-node test-server/test-scheduler.mjs
-```
-
-| Check                       | Expected               |
-| --------------------------- | ---------------------- |
-| `scheduler.active`          | `true`, 3 jobs         |
-| All jobs `lastResult`       | `"success"` after wait |
-| All jobs `lastError`        | `null`                 |
-| backup `runCount`           | ≥ 2                    |
-| vacuum + rebuild `runCount` | ≥ 1 each               |
-
----
-
-## Test Execution Order
-
-0. **Phase 0**: Seed Test Data (creates entries for FTS5, filter, and semantic search tests)
-1. **Phase 1**: Infrastructure (must pass before proceeding)
-2. **Phase 2**: Entry CRUD (creates additional test data + validates CRUD operations)
-3. **Phase 3**: Search + Analytics (requires entries from Phase 0 + 2)
-4. **Phase 4**: Relationships (links entries from Phase 0 + 2)
-5. **Phase 5**: Admin/Backup (test last to avoid data loss)
-6. **Phase 6**: Automated Scheduler (HTTP only — manual terminal test)
-
----
-
-## Success Criteria
-
-### Core Functionality
-
-- [ ] All 24 core tools execute without errors on happy paths (GitHub tools tested in `test-tools2.md`, team tools in `test-tools-team.md`)
-- [ ] Server instructions length respects `--instruction-level`: essential (~1.5K tokens) < standard (~1.7K) < full (~2.7K)
-- [ ] 45 core/local tools have `openWorldHint: false`; 16 GitHub tools have `openWorldHint: true` (61 total, 0 missing)
-
-### Entry CRUD
-
-- [ ] `create_entry` persists all optional fields: PR fields, workflow fields, `projectOwner`, `autoContext`
-- [ ] `create_entry` with `share_with_team: true` creates entries in both personal and team DBs
-- [ ] `create_entry` rejects invalid `entry_type` and `significance_type` with structured errors (not raw throws)
-- [ ] `create_entry` with `issue_number` auto-populates `issueUrl` from cached repo info
-- [ ] `get_entry_by_id` returns `importance` score (0.0-1.0) and `importanceBreakdown`
-- [ ] `get_recent_entries` with `is_personal` filter returns only matching entries
-- [ ] `update_entry` returns `success: false` for nonexistent entry IDs
-- [ ] `delete_entry` returns `success: false` for nonexistent entry IDs
-
-### Search & Analytics
-
-- [ ] `search_entries` FTS5 phrase, prefix, and boolean operators work correctly
-- [ ] `search_entries` gracefully falls back to LIKE for FTS5-unsafe queries (single quotes, `%`)
-- [ ] `search_entries` filters work: `issue_number`, `pr_status`, `workflow_run_id`, `project_number`, `is_personal`
-- [ ] `search_by_date_range` filters work: `entry_type`, `tags`, `is_personal`, `project_number`
-- [ ] `search_by_date_range` rejects non-YYYY-MM-DD date strings with structured errors
-- [ ] Cross-DB merging includes `source: 'personal' | 'team'` marker
-- [ ] `semantic_search` with custom `similarity_threshold` affects result count
-- [ ] `get_statistics` returns all 4 enhanced analytics metrics with correct groupings
-- [ ] `get_cross_project_insights` returns `project_count ≥ 1` with seed entries S7, S13, S14 present (project 5 has 3 entries, meeting `min_entries: 3`)
-- [ ] `get_cross_project_insights` response includes `top_tags`, `first_entry`, `last_entry`, `active_days`, `time_distribution` per project
-- [ ] `get_cross_project_insights` returns all required schema fields (including `projects: []`) when empty (tested with `min_entries: 9999`)
-
-### Relationships
-
-- [ ] Causal relationship types (`blocked_by`, `resolved`, `caused`) create correctly
-- [ ] `link_entries` returns `success: false` for nonexistent entry IDs
-- [ ] `visualize_relationships` shows distinct arrows for causal types
-- [ ] `memory://graph/recent` uses harmonized arrows matching `visualize_relationships`
-
-### Admin & Backup
-
-- [ ] `merge_tags` consolidates duplicate tags correctly — verified via `list_tags` and entry re-check
-- [ ] `merge_tags` returns structured error when source equals target or source tag nonexistent
-- [ ] `backup_journal` rejects names containing path traversal characters (`../`) with structured errors
-- [ ] `restore_backup` with nonexistent filename returns structured error
-
-### Scheduler (Phase 6)
-
-- [ ] `memory://health` shows `scheduler.active: false` and empty `jobs` array in stdio mode
-- [ ] All 3 jobs active with `nextRun` timestamps in HTTP mode
-- [ ] All `lastResult` values are `"success"` after jobs fire
-- [ ] Error in one job does not prevent others from running