ultimate-gemini-mcp 1.0.16__tar.gz → 1.0.18__tar.gz

ultimate_gemini_mcp-1.0.18/.genkit/traces_idx/genkit.metadata

@@ -0,0 +1 @@
+{"version":"1.21.0"}

ultimate_gemini_mcp-1.0.18/FIX_VERIFICATION_TEST_PLAN.md

@@ -0,0 +1,601 @@
+# Fix Verification Test Plan
+
+**Purpose:** Verify that all critical fixes from commit `efa6a4d` are working correctly
+**Date Created:** October 25, 2025
+**Prerequisites:** MCP client must be restarted to load updated code
+
+---
+
+## Test Categories
+
+1. [Gemini Image Generation](#1-gemini-image-generation-critical)
+2. [Seed Parameter Handling](#2-seed-parameter-handling)
+3. [Person Generation Policy Warnings](#3-person-generation-policy-warnings)
+4. [Aspect Ratio Support (Gemini)](#4-aspect-ratio-support-gemini)
+5. [Regression Tests](#5-regression-tests)
+
+---
+
+## 1. Gemini Image Generation (CRITICAL)
+
+**Issue Fixed:** Gemini API was returning no image data due to missing `responseModalities` config
+
+### Test 1.1: Basic Gemini Generation
+
+**Objective:** Verify basic image generation works with default settings
+
+**Steps:**
+```python
+generate_image(
+    prompt="a red apple on a wooden table",
+    model="gemini-2.5-flash-image",
+    enhance_prompt=False
+)
+```
+
+**Expected Result:**
+- ✅ `success: true`
+- ✅ Image file created in `generated_images/`
+- ✅ File size > 0 bytes
+- ✅ Response includes `path` and `filename`
+- ✅ No error messages
+
+**Failure Criteria:**
+- ❌ Error: "No image data found in Gemini API response"
+- ❌ `success: false`
+
+---
+
+### Test 1.2: Gemini with Prompt Enhancement
+
+**Objective:** Verify prompt enhancement doesn't break generation
+
+**Steps:**
+```python
+generate_image(
+    prompt="a futuristic robot in a neon city",
+    model="gemini-2.5-flash-image",
+    enhance_prompt=True
+)
+```
+
+**Expected Result:**
+- ✅ `success: true`
+- ✅ Image generated successfully
+- ✅ Response includes `enhanced_prompt` in metadata
+- ✅ Enhanced prompt is significantly longer than original
+
+**Failure Criteria:**
+- ❌ Enhancement breaks image generation
+- ❌ No enhanced_prompt in response
+
+---
+
+### Test 1.3: Gemini Image Editing
+
+**Objective:** Verify image editing with input image works
+
+**Prerequisites:** Input image file exists at root directory
+
+**Steps:**
+```python
+generate_image(
+    prompt="add a red border around this image",
+    model="gemini-2.5-flash-image",
+    input_image_path="/Users/nikhilanand/Desktop/ultimate gemini mcp/ultimate-gemini-mcp/thisimage.png"
+)
+```
+
+**Expected Result:**
+- ✅ `success: true`
+- ✅ Output image is modified version of input
+- ✅ Red border visible in output
+
+**Failure Criteria:**
+- ❌ Error: "No image data found"
+- ❌ Original image unchanged
+
+---
+
+### Test 1.4: Gemini Complex Prompt
+
+**Objective:** Verify Gemini handles complex, detailed prompts
+
+**Steps:**
+```python
+generate_image(
+    prompt="A photorealistic portrait of a wise old wizard with a long white beard, wearing a purple robe with gold embroidery, holding a crystal staff, dramatic lighting from the left side, mystical atmosphere, high detail",
+    model="gemini-2.5-flash-image",
+    enhance_prompt=True
+)
+```
+
+**Expected Result:**
+- ✅ `success: true`
+- ✅ Image reflects detailed prompt elements
+- ✅ Enhancement adds cinematic/photographic terms
+
+**Failure Criteria:**
+- ❌ Generic or low-quality output
+- ❌ Generation fails
+
+---
+
+## 2. Seed Parameter Handling
+
+**Issue Fixed:** Imagen API returns 400 error when seed is provided; now handled gracefully
+
+### Test 2.1: Seed Parameter with Warning
+
+**Objective:** Verify seed parameter triggers warning but doesn't fail
+
+**Steps:**
+```python
+generate_image(
+    prompt="a colorful abstract pattern",
+    model="imagen-4",
+    seed=12345
+)
+```
+
+**Expected Result:**
+- ✅ `success: true`
+- ✅ Image generated successfully
+- ✅ Warning logged: "seed parameter is not currently supported"
+- ✅ Seed is NOT sent to API (no 400 error)
+
+**Failure Criteria:**
+- ❌ Error: "Setting seed is not supported" (400 from API)
+- ❌ Generation fails
+- ❌ No warning message
+
+---
+
+### Test 2.2: Same Prompt Without Seed
+
+**Objective:** Verify removal of seed doesn't affect normal operation
+
+**Steps:**
+```python
+generate_image(
+    prompt="a colorful abstract pattern",
+    model="imagen-4",
+    seed=None  # or omit parameter
+)
+```
+
+**Expected Result:**
+- ✅ `success: true`
+- ✅ Image generated successfully
+- ✅ No warning about seed
+
+**Failure Criteria:**
+- ❌ Any errors
+- ❌ Unexpected warnings
+
+---
+
+### Test 2.3: Seed Documentation Check
+
+**Objective:** Verify tool documentation reflects seed limitation
+
+**Steps:**
+- Inspect the tool's docstring/help text
+
+**Expected Result:**
+- ✅ Seed parameter description includes: "NOT SUPPORTED - will be ignored"
+- ✅ Clear indication that seed won't work
+
+**Failure Criteria:**
+- ❌ Documentation claims seed works for reproducibility
+- ❌ No warning about limitation
+
+---
+
+## 3. Person Generation Policy Warnings
+
+**Issue Fixed:** Added validation to warn when prompt conflicts with person_generation policy
+
+### Test 3.1: Conflict Detection - "dont_allow" with Person Keywords
+
+**Objective:** Verify warning when prompt mentions people with restrictive policy
+
+**Steps:**
+```python
+generate_image(
+    prompt="a crowd of people walking in a busy street",
+    model="imagen-4",
+    person_generation="dont_allow"
+)
+```
+
+**Expected Result:**
+- ✅ Warning logged: "Prompt contains person-related keywords but person_generation is set to 'dont_allow'"
+- ✅ Warning mentions: "may result in the API blocking image generation"
+- ⚠️ May succeed or fail depending on API behavior
+- ✅ User is warned BEFORE API call
+
+**Failure Criteria:**
+- ❌ No warning issued
+- ❌ Silent failure without explanation
+
+---
+
+### Test 3.2: Conflict Detection - Multiple Person Keywords
+
+**Objective:** Verify detection of various person-related terms
+
+**Test Cases:**
+| Prompt | Should Warn |
+|--------|-------------|
+| "a portrait of a woman" | ✅ Yes |
+| "human face close-up" | ✅ Yes |
+| "child playing in park" | ✅ Yes |
+| "man standing alone" | ✅ Yes |
+| "a landscape with trees" | ❌ No |
+| "abstract geometric shapes" | ❌ No |
+
+**Steps:**
+Test each prompt with `person_generation="dont_allow"`
+
+**Expected Result:**
+- ✅ Warnings issued only for prompts with person keywords
+- ✅ No false positives on landscape/abstract prompts
+
+---
+
+### Test 3.3: No Warning with "allow_adult"
+
+**Objective:** Verify no warning when policy allows people
+
+**Steps:**
+```python
+generate_image(
+    prompt="a person walking in a park",
+    model="imagen-4",
+    person_generation="allow_adult"  # default
+)
+```
+
+**Expected Result:**
+- ✅ `success: true`
+- ✅ Image generated with person
+- ✅ No conflict warning (policy allows people)
+
+**Failure Criteria:**
+- ❌ Warning issued despite permissive policy
+- ❌ Generation blocked
+
+---
+
+## 4. Aspect Ratio Support (Gemini)
+
+**Issue Fixed:** Added `imageConfig` with `aspectRatio` to Gemini requests
+
+### Test 4.1: Gemini 1:1 (Default)
+
+**Steps:**
+```python
+generate_image(
+    prompt="a simple geometric shape",
+    model="gemini-2.5-flash-image",
+    aspect_ratio="1:1"
+)
+```
+
+**Expected Result:**
+- ✅ Image is square (1:1 ratio)
+- ✅ `imageConfig.aspectRatio: "1:1"` sent to API
+
+---
+
+### Test 4.2: Gemini 16:9 (Wide)
+
+**Steps:**
+```python
+generate_image(
+    prompt="a panoramic mountain landscape",
+    model="gemini-2.5-flash-image",
+    aspect_ratio="16:9"
+)
+```
+
+**Expected Result:**
+- ✅ Image is wide (16:9 ratio)
+- ✅ Landscape composition utilized
+
+---
+
+### Test 4.3: Gemini 9:16 (Tall)
+
+**Steps:**
+```python
+generate_image(
+    prompt="a tall skyscraper reaching into clouds",
+    model="gemini-2.5-flash-image",
+    aspect_ratio="9:16"
+)
+```
+
+**Expected Result:**
+- ✅ Image is tall/vertical (9:16 ratio)
+- ✅ Composition emphasizes height
+
+---
+
+### Test 4.4: Gemini Multiple Aspect Ratios
+
+**Objective:** Test all supported aspect ratios
+
+**Test Matrix:**
+| Aspect Ratio | Expected Orientation | Test Prompt |
+|--------------|---------------------|-------------|
+| 1:1 | Square | "a perfect circle" |
+| 2:3 | Portrait | "a vertical portrait" |
+| 3:2 | Landscape | "a wide landscape" |
+| 4:3 | Classic horizontal | "a classic photo" |
+| 16:9 | Widescreen | "an ultra-wide panorama" |
+| 9:16 | Mobile portrait | "a tall mobile wallpaper" |
+
+**Expected Result:**
+- ✅ All ratios generate successfully
+- ✅ Output images match requested ratios
+
+---
+
+## 5. Regression Tests
+
+**Objective:** Ensure fixes didn't break existing functionality
+
+### Test 5.1: Imagen Still Works
+
+**Steps:**
+```python
+generate_image(
+    prompt="a beautiful sunset",
+    model="imagen-4"
+)
+```
+
+**Expected Result:**
+- ✅ Imagen generation still works perfectly
+- ✅ No changes to Imagen behavior
+
+---
+
+### Test 5.2: Negative Prompt Still Works
+
+**Steps:**
+```python
+generate_image(
+    prompt="a forest landscape",
+    model="imagen-4",
+    negative_prompt="people, buildings, cars"
+)
+```
+
+**Expected Result:**
+- ✅ Negative prompts still applied correctly
+- ✅ Excluded elements don't appear
+
+---
+
+### Test 5.3: Prompt Enhancement Still Works
+
+**Steps:**
+```python
+generate_image(
+    prompt="a simple scene",
+    model="imagen-4",
+    enhance_prompt=True
+)
+```
+
+**Expected Result:**
+- ✅ Enhancement still adds detailed photography terms
+- ✅ Enhanced prompt visible in response metadata
+
+---
+
+### Test 5.4: Input Validation Still Works
+
+**Steps:**
+```python
+# Test invalid model
+generate_image(
+    prompt="test",
+    model="invalid-model-999"
+)
+
+# Test invalid aspect ratio
+generate_image(
+    prompt="test",
+    model="imagen-4",
+    aspect_ratio="999:1"
+)
+```
+
+**Expected Result:**
+- ✅ Validation errors for invalid inputs
+- ✅ Clear error messages with available options
+
+---
+
+## Debug Logging Verification
+
+**Issue Fixed:** Added comprehensive debug logging
+
+### Test 6.1: Debug Logs Present (if LOG_LEVEL=DEBUG)
+
+**Steps:**
+1. Set environment variable: `LOG_LEVEL=DEBUG`
+2. Run any Gemini generation
+3. Check logs
+
+**Expected Log Entries:**
+- ✅ "Sending request to [url]"
+- ✅ "Request body: {...}"
+- ✅ "Response status: 200"
+- ✅ "Response data: {...}"
+- ✅ "Extracted image data of length: [bytes]"
+
+**Failure Criteria:**
+- ❌ No debug logs visible
+- ❌ Missing request/response details
+
+---
+
+### Test 6.2: Error Logging Improvements
+
+**Steps:**
+Force an error (e.g., invalid API key) and check logs
+
+**Expected Log Entries:**
+- ✅ "No images extracted from response. Response structure: [keys]"
+- ✅ "Candidates: [candidate data]" (if candidates present)
+- ✅ Detailed error context
+
+---
+
+## Test Execution Checklist
+
+### Pre-Test Setup
+- [ ] MCP client restarted (to load new code)
+- [ ] `GEMINI_API_KEY` environment variable set
+- [ ] Input images available at root directory (`thisimage.png`, `thisOnetoo.png`)
+- [ ] `generated_images/` directory exists
+
+### Critical Tests (Must Pass)
+- [ ] Test 1.1: Basic Gemini Generation ⭐
+- [ ] Test 1.2: Gemini with Prompt Enhancement ⭐
+- [ ] Test 1.3: Gemini Image Editing ⭐
+- [ ] Test 2.1: Seed Parameter with Warning ⭐
+- [ ] Test 3.1: Person Policy Conflict Warning ⭐
+- [ ] Test 4.2: Gemini 16:9 Aspect Ratio ⭐
+
+### Important Tests (Should Pass)
+- [ ] Test 1.4: Complex Prompt
+- [ ] Test 2.3: Seed Documentation Check
+- [ ] Test 3.2: Multiple Person Keywords
+- [ ] Test 4.4: All Aspect Ratios
+
+### Regression Tests (Must Pass)
+- [ ] Test 5.1: Imagen Still Works
+- [ ] Test 5.2: Negative Prompt Still Works
+- [ ] Test 5.3: Prompt Enhancement Still Works
+- [ ] Test 5.4: Input Validation Still Works
+
+### Optional Tests
+- [ ] Test 6.1: Debug Logging
+- [ ] Test 6.2: Error Logging
+
+---
+
+## Success Criteria
+
+### Minimum Acceptable (MVP)
+- ✅ All 6 Critical Tests pass
+- ✅ At least 3 out of 4 Important Tests pass
+- ✅ All 4 Regression Tests pass
+- ✅ Zero P0 (critical) bugs found
+
+### Full Success
+- ✅ All Critical Tests pass (100%)
+- ✅ All Important Tests pass (100%)
+- ✅ All Regression Tests pass (100%)
+- ✅ Optional Tests pass (bonus)
+- ✅ No bugs found
+
+### Partial Success
+- ⚠️ 4-5 Critical Tests pass (80%+)
+- ⚠️ Most Regression Tests pass
+- ⚠️ New issues documented for follow-up
+
+---
+
+## Issue Tracking Template
+
+If tests fail, document using this template:
+
+```markdown
+### Issue: [Brief Description]
+
+**Test Failed:** [Test Number and Name]
+**Severity:** Critical | High | Medium | Low
+**Status:** New | In Progress | Fixed
+
+**Steps to Reproduce:**
+1.
+2.
+3.
+
+**Expected Behavior:**
+[What should happen]
+
+**Actual Behavior:**
+[What actually happened]
+
+**Error Messages:**
+```
+[Paste error output]
+```
+
+**Additional Context:**
+[Logs, screenshots, etc.]
+
+**Proposed Fix:**
+[If known]
+```
+
+---
+
+## Post-Test Actions
+
+### If All Tests Pass ✅
+1. Update `TEST_RESULTS.md` with verification results
+2. Mark issues as "FIXED" in documentation
+3. Update project status to "Stable"
+4. Consider bumping version to v1.1.0
+
+### If Tests Fail ❌
+1. Document failures using Issue Tracking Template
+2. Prioritize by severity
+3. Create fix plan for critical issues
+4. Re-run tests after fixes
+
+### Partial Success ⚠️
+1. Document which tests passed/failed
+2. Assess if partial fix is better than no fix
+3. Decide: rollback or proceed with known issues
+4. Update documentation with limitations
+
+---
+
+## Notes
+
+- **Test Execution Order:** Run Critical tests first, then Regression, then Important
+- **Parallel Testing:** Regression tests can run in parallel with fix verification
+- **Time Estimate:** ~20-30 minutes for full test suite
+- **Quick Verification:** Run Critical Tests only (~5 minutes)
+
+---
+
+## Appendix: Quick Test Commands
+
+### Minimal Verification (2 minutes)
+```python
+# Test 1: Gemini works
+generate_image(prompt="a red apple", model="gemini-2.5-flash-image", enhance_prompt=False)
+
+# Test 2: Seed handled gracefully
+generate_image(prompt="abstract art", model="imagen-4", seed=123)
+```
+
+### Full Verification Script
+See `run_verification_tests.py` (to be created) for automated testing.
+
+---
+
+**Document Version:** 1.0
+**Last Updated:** October 25, 2025
+**Next Review:** After test execution

{ultimate_gemini_mcp-1.0.16 → ultimate_gemini_mcp-1.0.18}/PKG-INFO

@@ -1,6 +1,6 @@
 Metadata-Version: 2.4
 Name: ultimate-gemini-mcp
-Version: 1.0.16
+Version: 1.0.18
 Summary: Ultimate image generation MCP server unifying Gemini 2.5 Flash Image and Imagen 4/Fast/Ultra with advanced features
 Project-URL: Homepage, https://github.com/anand-92/ultimate-image-gen-mcp
 Project-URL: Repository, https://github.com/anand-92/ultimate-image-gen-mcp

{ultimate_gemini_mcp-1.0.16 → ultimate_gemini_mcp-1.0.18}/pyproject.toml

@@ -1,6 +1,6 @@
 [project]
 name = "ultimate-gemini-mcp"
-version = "1.0.16"
+version = "1.0.18"
 description = "Ultimate image generation MCP server unifying Gemini 2.5 Flash Image and Imagen 4/Fast/Ultra with advanced features"
 readme = "README.md"
 requires-python = ">=3.11"