verisight 2.2.1__tar.gz → 2.2.3__tar.gz

This diff represents the content of publicly available package versions that have been released to one of the supported registries. The information contained in this diff is provided for informational purposes only and reflects changes between package versions as they appear in their respective public registries.
Files changed (23) hide show
  1. {verisight-2.2.1/verisight.egg-info → verisight-2.2.3}/PKG-INFO +30 -2
  2. {verisight-2.2.1 → verisight-2.2.3}/README.md +29 -1
  3. {verisight-2.2.1 → verisight-2.2.3}/backend/__init__.py +1 -1
  4. {verisight-2.2.1 → verisight-2.2.3}/backend/agent.py +49 -27
  5. {verisight-2.2.1 → verisight-2.2.3}/backend/main.py +4 -4
  6. {verisight-2.2.1 → verisight-2.2.3}/backend/mcp_server.py +2 -2
  7. {verisight-2.2.1 → verisight-2.2.3}/pyproject.toml +1 -1
  8. {verisight-2.2.1 → verisight-2.2.3/verisight.egg-info}/PKG-INFO +30 -2
  9. {verisight-2.2.1 → verisight-2.2.3}/LICENSE +0 -0
  10. {verisight-2.2.1 → verisight-2.2.3}/backend/aws_adapter.py +0 -0
  11. {verisight-2.2.1 → verisight-2.2.3}/backend/benchmark_runner.py +0 -0
  12. {verisight-2.2.1 → verisight-2.2.3}/backend/cli.py +0 -0
  13. {verisight-2.2.1 → verisight-2.2.3}/backend/forensics.py +0 -0
  14. {verisight-2.2.1 → verisight-2.2.3}/backend/generate_demo_samples.py +0 -0
  15. {verisight-2.2.1 → verisight-2.2.3}/backend/generate_real_life_samples.py +0 -0
  16. {verisight-2.2.1 → verisight-2.2.3}/backend/test_mcp_tools.py +0 -0
  17. {verisight-2.2.1 → verisight-2.2.3}/backend/test_sanity.py +0 -0
  18. {verisight-2.2.1 → verisight-2.2.3}/setup.cfg +0 -0
  19. {verisight-2.2.1 → verisight-2.2.3}/verisight.egg-info/SOURCES.txt +0 -0
  20. {verisight-2.2.1 → verisight-2.2.3}/verisight.egg-info/dependency_links.txt +0 -0
  21. {verisight-2.2.1 → verisight-2.2.3}/verisight.egg-info/entry_points.txt +0 -0
  22. {verisight-2.2.1 → verisight-2.2.3}/verisight.egg-info/requires.txt +0 -0
  23. {verisight-2.2.1 → verisight-2.2.3}/verisight.egg-info/top_level.txt +0 -0
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: verisight
3
- Version: 2.2.1
3
+ Version: 2.2.3
4
4
  Summary: Sub-3ms Physical Multi-Spectral Image Forensics & Presentation Attack Detection engine powered by OpenCV 5 and Model Context Protocol (MCP).
5
5
  Author-email: Aditya Raj Tiwari <adityasantosh2026@gmail.com>
6
6
  License-Expression: Apache-2.0
@@ -34,6 +34,8 @@ Dynamic: license-file
34
34
  # 🔬 VeriSight MCP — Autonomous Physical AI Visual Forensics Engine
35
35
 
36
36
  [![VeriSight CI](https://github.com/Adityarajtiwari880/verisight/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/Adityarajtiwari880/verisight/actions/workflows/ci.yml)
37
+ [![PyPI version](https://img.shields.io/pypi/v/verisight.svg?color=blue)](https://pypi.org/project/verisight/)
38
+ [![mcprush score](https://mcprush.com/badge/verisight-mcp/score.svg)](https://mcprush.com/to/verisight-mcp?ref=badge)
37
39
  [![M8ven Score](https://m8ven.ai/badge/mcp/adityarajtiwari880-verisight-11ohzr?v=5ddb567da1a0cb2e2f12567300d7ff24)](https://m8ven.ai/mcp/adityarajtiwari880-verisight-11ohzr?s=readme)
38
40
  [![License](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)
39
41
  [![Python Version](https://img.shields.io/badge/python-3.10%20%7C%203.11%20%7C%203.12-3776AB?logo=python&logoColor=white)](pyproject.toml)
@@ -43,6 +45,10 @@ Dynamic: license-file
43
45
  [![Next.js 16](https://img.shields.io/badge/Next.js-16.4.3-000000?logo=next.js&logoColor=white)](https://nextjs.org/)
44
46
  [![FastAPI](https://img.shields.io/badge/FastAPI-Production%20Ready-009688?logo=fastapi&logoColor=white)](https://fastapi.tiangolo.com/)
45
47
 
48
+ <p align="center">
49
+ <a href="https://mcprush.com/to/verisight-mcp?ref=badge"><img width="380" height="200" src="https://mcprush.com/badge/verisight-mcp/card.svg" alt="VeriSight MCP server" /></a>
50
+ </p>
51
+
46
52
  > **Sub-3ms Physical AI and multi-spectral visual forensics engine built on pure OpenCV 5 and SIMD acceleration.** Detects generative AI inpainting, deepfake manipulation, duplicate photo reuse, and presentation recapture attacks with zero heavyweight neural network dependencies.
47
53
 
48
54
  ---
@@ -77,6 +83,23 @@ Authentic Image ──> Generative AI Inpaint (Flux / Firefly) ──> Instant R
77
83
  | **Generalization & Drift** | Vulnerable to adversarial attacks & new diffusion models | **Grounded in mathematical physical optics & sensor physics** |
78
84
  | **Audit Explainability** | Black-box softmax probabilities | **Deterministic mathematical proof (ELA, PRNU, 2D FFT, Ghosts)** |
79
85
  | **Recapture Attack Defense**| None (Foiled by 2D screen or paper printout) | **Built-in Presentation Attack Detection (PAD)** |
86
+ | **External API Cost** | \$0.01–\$0.05/image via VLM/SaaS wrappers | **\$0.00 (100% offline/local C++ OpenCV 5)** |
87
+
88
+ ---
89
+
90
+ ## 🎯 Verified Real-World Evaluation Suite (`10/10` Passing in CI)
91
+
92
+ Verified automatically on every commit via [`backend/test_sanity.py`](backend/test_sanity.py) and [`backend/test_mcp_tools.py`](backend/test_mcp_tools.py):
93
+
94
+ | Test Scenario | Ground Truth | VeriSight Verdict | Risk Score |
95
+ | :--- | :--- | :--- | :---: |
96
+ | Genuine Unedited Photo | Authentic | `AUTO_APPROVE_CLAIM` | `10.9%` |
97
+ | Generative AI Inpainted Damage | Fraud | `REJECT_SYNTHETIC_TAMPER` | `99.4%` |
98
+ | Genuine Digital Coupon Screenshot | Authentic | `AUTO_APPROVE_CLAIM` | `30.8%` |
99
+ | AI Altered Coupon Code (`D` → `C`) | Fraud | `REJECT_SYNTHETIC_TAMPER` | `99.4%` |
100
+ | Printed Card Recapture Attack | PAD Attack | `REJECT_RECAPTURED_2D_MEDIA` | `97.5%` |
101
+ | Duplicate Photo Reuse (Sybil Ring) | Reuse Fraud | `REJECT_DUPLICATE_CLAIM_FRAUD` | `99.4%` |
102
+ | Selfie / Off-Subject in Produce or Invoice | Invalid Subject | `REJECT_INVALID_CLAIM_SUBJECT` | `Blocked` |
80
103
 
81
104
  ---
82
105
 
@@ -148,12 +171,17 @@ verisight server --port 8000
148
171
 
149
172
  VeriSight natively implements the **Anthropic Model Context Protocol (MCP)** specification (`protocol_version: 2024-11-05`). External LLM agents (**Claude Desktop**, **Cursor**, **Windsurf**, **LangChain**) can discover and call VeriSight tools autonomously via high-performance standard I/O (`stdio`).
150
173
 
151
- ### 1-Click Zero-Install (`uvx`)
174
+ ### 1-Click Zero-Install (`uvx` & `claude mcp`)
152
175
  Run directly via Astral `uvx` with zero manual cloning or environment setup:
153
176
  ```bash
154
177
  uvx verisight mcp
155
178
  ```
156
179
 
180
+ Or add directly to **Claude Code CLI**:
181
+ ```bash
182
+ claude mcp add --scope user verisight-mcp '--' uvx verisight mcp
183
+ ```
184
+
157
185
  ### Claude Desktop & Cursor Configuration
158
186
 
159
187
  Add VeriSight directly to your `claude_desktop_config.json` (or Cursor MCP settings):
@@ -1,6 +1,8 @@
1
1
  # 🔬 VeriSight MCP — Autonomous Physical AI Visual Forensics Engine
2
2
 
3
3
  [![VeriSight CI](https://github.com/Adityarajtiwari880/verisight/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/Adityarajtiwari880/verisight/actions/workflows/ci.yml)
4
+ [![PyPI version](https://img.shields.io/pypi/v/verisight.svg?color=blue)](https://pypi.org/project/verisight/)
5
+ [![mcprush score](https://mcprush.com/badge/verisight-mcp/score.svg)](https://mcprush.com/to/verisight-mcp?ref=badge)
4
6
  [![M8ven Score](https://m8ven.ai/badge/mcp/adityarajtiwari880-verisight-11ohzr?v=5ddb567da1a0cb2e2f12567300d7ff24)](https://m8ven.ai/mcp/adityarajtiwari880-verisight-11ohzr?s=readme)
5
7
  [![License](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)
6
8
  [![Python Version](https://img.shields.io/badge/python-3.10%20%7C%203.11%20%7C%203.12-3776AB?logo=python&logoColor=white)](pyproject.toml)
@@ -10,6 +12,10 @@
10
12
  [![Next.js 16](https://img.shields.io/badge/Next.js-16.4.3-000000?logo=next.js&logoColor=white)](https://nextjs.org/)
11
13
  [![FastAPI](https://img.shields.io/badge/FastAPI-Production%20Ready-009688?logo=fastapi&logoColor=white)](https://fastapi.tiangolo.com/)
12
14
 
15
+ <p align="center">
16
+ <a href="https://mcprush.com/to/verisight-mcp?ref=badge"><img width="380" height="200" src="https://mcprush.com/badge/verisight-mcp/card.svg" alt="VeriSight MCP server" /></a>
17
+ </p>
18
+
13
19
  > **Sub-3ms Physical AI and multi-spectral visual forensics engine built on pure OpenCV 5 and SIMD acceleration.** Detects generative AI inpainting, deepfake manipulation, duplicate photo reuse, and presentation recapture attacks with zero heavyweight neural network dependencies.
14
20
 
15
21
  ---
@@ -44,6 +50,23 @@ Authentic Image ──> Generative AI Inpaint (Flux / Firefly) ──> Instant R
44
50
  | **Generalization & Drift** | Vulnerable to adversarial attacks & new diffusion models | **Grounded in mathematical physical optics & sensor physics** |
45
51
  | **Audit Explainability** | Black-box softmax probabilities | **Deterministic mathematical proof (ELA, PRNU, 2D FFT, Ghosts)** |
46
52
  | **Recapture Attack Defense**| None (Foiled by 2D screen or paper printout) | **Built-in Presentation Attack Detection (PAD)** |
53
+ | **External API Cost** | \$0.01–\$0.05/image via VLM/SaaS wrappers | **\$0.00 (100% offline/local C++ OpenCV 5)** |
54
+
55
+ ---
56
+
57
+ ## 🎯 Verified Real-World Evaluation Suite (`10/10` Passing in CI)
58
+
59
+ Verified automatically on every commit via [`backend/test_sanity.py`](backend/test_sanity.py) and [`backend/test_mcp_tools.py`](backend/test_mcp_tools.py):
60
+
61
+ | Test Scenario | Ground Truth | VeriSight Verdict | Risk Score |
62
+ | :--- | :--- | :--- | :---: |
63
+ | Genuine Unedited Photo | Authentic | `AUTO_APPROVE_CLAIM` | `10.9%` |
64
+ | Generative AI Inpainted Damage | Fraud | `REJECT_SYNTHETIC_TAMPER` | `99.4%` |
65
+ | Genuine Digital Coupon Screenshot | Authentic | `AUTO_APPROVE_CLAIM` | `30.8%` |
66
+ | AI Altered Coupon Code (`D` → `C`) | Fraud | `REJECT_SYNTHETIC_TAMPER` | `99.4%` |
67
+ | Printed Card Recapture Attack | PAD Attack | `REJECT_RECAPTURED_2D_MEDIA` | `97.5%` |
68
+ | Duplicate Photo Reuse (Sybil Ring) | Reuse Fraud | `REJECT_DUPLICATE_CLAIM_FRAUD` | `99.4%` |
69
+ | Selfie / Off-Subject in Produce or Invoice | Invalid Subject | `REJECT_INVALID_CLAIM_SUBJECT` | `Blocked` |
47
70
 
48
71
  ---
49
72
 
@@ -115,12 +138,17 @@ verisight server --port 8000
115
138
 
116
139
  VeriSight natively implements the **Anthropic Model Context Protocol (MCP)** specification (`protocol_version: 2024-11-05`). External LLM agents (**Claude Desktop**, **Cursor**, **Windsurf**, **LangChain**) can discover and call VeriSight tools autonomously via high-performance standard I/O (`stdio`).
117
140
 
118
- ### 1-Click Zero-Install (`uvx`)
141
+ ### 1-Click Zero-Install (`uvx` & `claude mcp`)
119
142
  Run directly via Astral `uvx` with zero manual cloning or environment setup:
120
143
  ```bash
121
144
  uvx verisight mcp
122
145
  ```
123
146
 
147
+ Or add directly to **Claude Code CLI**:
148
+ ```bash
149
+ claude mcp add --scope user verisight-mcp '--' uvx verisight mcp
150
+ ```
151
+
124
152
  ### Claude Desktop & Cursor Configuration
125
153
 
126
154
  Add VeriSight directly to your `claude_desktop_config.json` (or Cursor MCP settings):
@@ -30,7 +30,7 @@ from forensics import (
30
30
  tool_perceptual_hash
31
31
  )
32
32
 
33
- __version__ = "2.2.1"
33
+ __version__ = "2.2.3"
34
34
  __author__ = "Aditya Raj Tiwari"
35
35
  __license__ = "Apache-2.0"
36
36
 
@@ -38,7 +38,7 @@ class ForensicAgent:
38
38
  JPEG Ghost Sweeper, and PRNU Sensor Noise.
39
39
  """
40
40
  def __init__(self):
41
- self.version = "2.2.1"
41
+ self.version = "2.2.3"
42
42
 
43
43
  def run_investigation(
44
44
  self,
@@ -443,51 +443,73 @@ class ForensicAgent:
443
443
  is_full_frame = (rw > 0.65 * working_dim[0] and rh > 0.65 * working_dim[1])
444
444
  has_isolated_cluster = not is_full_frame and (clusters_detected > 0) and (ela_score > 25.0)
445
445
 
446
- # Generative AI detection:
447
- # Detects diffusion micro-fringing, synthetic sensor voids, or lattice harmonics
448
- is_generative_ai = (
449
- (cr_sobel_std >= 6.0 and ela_score >= 35.0) or
450
- (sensor_noise_std < 1.15 and ela_score >= 30.0) or
451
- (harmonic_ratio >= 2.8 and fft_energy >= 85.0 and ela_score >= 38.0)
446
+ # Natural Camera vs Document / Synthetic Domain Differentiation:
447
+ # In natural camera photos (damaged items, vehicles, nature, people), natural optical color transitions
448
+ # produce high whole-frame chrominance variance that must not be confused with generative AI fringing.
449
+ # In documents/vouchers, paper is predominantly monochrome/duotone so character editing creates high chrominance spikes.
450
+ is_doc_or_voucher = scene_triage.get("has_document_structure", False) or any(
451
+ k in claim_type.lower() for k in ["voucher", "coupon", "bill", "invoice", "receipt", "document", "text"]
452
452
  )
453
453
 
454
- # Multi-modal concordance: Requires corroboration from multiple orthogonal forensic probes
455
- has_corroboration = (
456
- is_feathered or
457
- (ghost_discrepancy >= 0.8) or
458
- (noise_divergence >= 45.0 and ela_score >= 25.0) or
459
- (ela_score >= 68.0) or
460
- is_generative_ai or
461
- provenance.get("is_edited_software_flag", False)
462
- )
454
+ # Generative AI detection & Corroboration:
455
+ if is_doc_or_voucher:
456
+ is_generative_ai = (
457
+ (cr_sobel_std >= 6.0 and ela_score >= 35.0 and has_isolated_cluster) or
458
+ (sensor_noise_std < 1.15 and ela_score >= 30.0) or
459
+ (harmonic_ratio >= 2.8 and fft_energy >= 85.0 and ela_score >= 38.0) or
460
+ (ghost_discrepancy >= 0.82 and noise_divergence >= 35.0 and has_isolated_cluster) or
461
+ (is_feathered and has_isolated_cluster)
462
+ )
463
+ has_corroboration = (
464
+ is_feathered or
465
+ (ghost_discrepancy >= 0.8) or
466
+ (noise_divergence >= 45.0 and ela_score >= 25.0) or
467
+ (ela_score >= 68.0) or
468
+ is_generative_ai or
469
+ provenance.get("is_edited_software_flag", False)
470
+ )
471
+ else:
472
+ is_generative_ai = (
473
+ (has_isolated_cluster and is_feathered and ela_score >= 35.0) or
474
+ (has_isolated_cluster and sensor_noise_std < 1.15 and noise_divergence >= 35.0 and ela_score >= 30.0) or
475
+ (harmonic_ratio >= 2.8 and fft_energy >= 85.0 and is_feathered) or
476
+ (has_isolated_cluster and is_feathered and (ghost_discrepancy >= 0.82 or noise_divergence >= 40.0))
477
+ )
478
+ has_corroboration = (
479
+ is_feathered or
480
+ (ghost_discrepancy >= 0.82 and has_isolated_cluster and is_feathered) or
481
+ (noise_divergence >= 45.0 and ela_score >= 35.0 and has_isolated_cluster and is_feathered) or
482
+ is_generative_ai or
483
+ provenance.get("is_edited_software_flag", False)
484
+ )
463
485
 
464
486
  # 1. Calibrated ELA contribution:
465
487
  # Organic cameras: ELA <= 20 (contrib 0-5%)
466
- # Moderate recompression: ELA 20-35 (contrib 5-32%)
467
- # Corroborated inpainting: ELA > 35 (contrib 32-75%)
468
- # Uncorroborated document text/graphics: capped at 18% to avoid false positives
488
+ # Moderate recompression: ELA 20-35 (contrib 5-27%)
489
+ # Corroborated inpainting: ELA > 35 (contrib 28-75%)
490
+ # Uncorroborated natural optical edges / texture: capped safely at 16%
469
491
  if ela_score <= 20.0:
470
492
  ela_contrib = max(0.0, (ela_score - 14.0) * 0.8)
471
493
  elif ela_score <= 35.0:
472
- ela_contrib = 5.0 + (ela_score - 20.0) * 1.8
494
+ ela_contrib = 5.0 + (ela_score - 20.0) * 1.5
473
495
  else:
474
- if has_corroboration:
475
- ela_contrib = 32.0 + (ela_score - 35.0) * 2.2
496
+ if has_corroboration and has_isolated_cluster:
497
+ ela_contrib = 28.0 + (ela_score - 35.0) * 1.8
476
498
  else:
477
- ela_contrib = min(18.0, 5.0 + (ela_score - 20.0) * 0.50)
499
+ ela_contrib = min(16.0, 5.0 + (ela_score - 20.0) * 0.35)
478
500
 
479
501
  # 2. Diffusion / Generative AI Spectral penalty:
480
502
  diffusion_contrib = 30.0 if is_generative_ai else 0.0
481
503
 
482
504
  # 3. Chrominance & Multi-probe contributions:
483
- if has_corroboration:
484
- chrom_contrib = min(40.0, max(0.0, (chrom_diff - 25.0) * 0.45)) if has_isolated_cluster else 0.0
505
+ if has_corroboration and has_isolated_cluster:
506
+ chrom_contrib = min(35.0, max(0.0, (chrom_diff - 25.0) * 0.45))
485
507
  else:
486
- chrom_contrib = min(6.0, max(0.0, (chrom_diff - 25.0) * 0.10)) if has_isolated_cluster else 0.0
508
+ chrom_contrib = min(5.0, max(0.0, (chrom_diff - 25.0) * 0.08)) if has_isolated_cluster else 0.0
487
509
 
488
510
  noise_contrib = max(0.0, (noise_divergence - 70.0) * 0.35) if (has_isolated_cluster and has_corroboration) else 0.0
489
511
  ghost_contrib = max(0.0, (ghost_discrepancy - 0.8) * 6.0) if (has_isolated_cluster and has_corroboration) else 0.0
490
- seam_contrib = 22.0 if (is_feathered and has_isolated_cluster) else 0.0
512
+ seam_contrib = 25.0 if (is_feathered and has_isolated_cluster) else 0.0
491
513
  software_contrib = 35.0 if provenance.get("is_edited_software_flag") else 0.0
492
514
 
493
515
  # Sensitivity modifier
@@ -31,7 +31,7 @@ from mcp_server import MCP_TOOLS_MANIFEST, handle_json_rpc
31
31
  app = FastAPI(
32
32
  title="VeriSight MCP - OpenCV 5 Agentic Forensics Engine",
33
33
  description="Multi-spectral forensic vision agent powered by OpenCV 5 on AWS Graviton Arm64",
34
- version="2.2.1"
34
+ version="2.2.3"
35
35
  )
36
36
 
37
37
  app.add_middleware(
@@ -47,7 +47,7 @@ def root_status():
47
47
  return {
48
48
  "service": "VeriSight MCP - Autonomous Physical AI Forensics Engine",
49
49
  "status": "ONLINE",
50
- "version": "2.2.1",
50
+ "version": "2.2.3",
51
51
  "engine": "OpenCV 5.0.0 (ARM64 SIMD)",
52
52
  "docs_url": "/docs",
53
53
  "api_url": "/api",
@@ -56,7 +56,7 @@ def root_status():
56
56
 
57
57
  @app.get("/health")
58
58
  def health_check():
59
- return {"status": "healthy", "service": "verisight-backend", "version": "2.2.1"}
59
+ return {"status": "healthy", "service": "verisight-backend", "version": "2.2.3"}
60
60
 
61
61
  @app.get("/api")
62
62
  @app.get("/api/")
@@ -64,7 +64,7 @@ def api_index():
64
64
  return {
65
65
  "service": "VeriSight MCP REST API",
66
66
  "status": "ONLINE",
67
- "version": "2.2.1",
67
+ "version": "2.2.3",
68
68
  "documentation": "/docs",
69
69
  "endpoints": {
70
70
  "health": "/health",
@@ -251,7 +251,7 @@ def handle_json_rpc(payload: Dict[str, Any], agent_pipeline_callback) -> Optiona
251
251
  },
252
252
  "serverInfo": {
253
253
  "name": "verisight-mcp-engine",
254
- "version": "2.2.1"
254
+ "version": "2.2.3"
255
255
  }
256
256
  }
257
257
  }
@@ -268,7 +268,7 @@ def handle_json_rpc(payload: Dict[str, Any], agent_pipeline_callback) -> Optiona
268
268
  "protocol_version": "2024-11-05",
269
269
  "server_info": {
270
270
  "name": "verisight-mcp-engine",
271
- "version": "2.2.1",
271
+ "version": "2.2.3",
272
272
  "capabilities": {"tools": True, "resources": False, "prompts": False}
273
273
  }
274
274
  }
@@ -4,7 +4,7 @@ build-backend = "setuptools.build_meta"
4
4
 
5
5
  [project]
6
6
  name = "verisight"
7
- version = "2.2.1"
7
+ version = "2.2.3"
8
8
  authors = [
9
9
  { name="Aditya Raj Tiwari", email="adityasantosh2026@gmail.com" },
10
10
  ]
@@ -1,6 +1,6 @@
1
1
  Metadata-Version: 2.4
2
2
  Name: verisight
3
- Version: 2.2.1
3
+ Version: 2.2.3
4
4
  Summary: Sub-3ms Physical Multi-Spectral Image Forensics & Presentation Attack Detection engine powered by OpenCV 5 and Model Context Protocol (MCP).
5
5
  Author-email: Aditya Raj Tiwari <adityasantosh2026@gmail.com>
6
6
  License-Expression: Apache-2.0
@@ -34,6 +34,8 @@ Dynamic: license-file
34
34
  # 🔬 VeriSight MCP — Autonomous Physical AI Visual Forensics Engine
35
35
 
36
36
  [![VeriSight CI](https://github.com/Adityarajtiwari880/verisight/actions/workflows/ci.yml/badge.svg?branch=main)](https://github.com/Adityarajtiwari880/verisight/actions/workflows/ci.yml)
37
+ [![PyPI version](https://img.shields.io/pypi/v/verisight.svg?color=blue)](https://pypi.org/project/verisight/)
38
+ [![mcprush score](https://mcprush.com/badge/verisight-mcp/score.svg)](https://mcprush.com/to/verisight-mcp?ref=badge)
37
39
  [![M8ven Score](https://m8ven.ai/badge/mcp/adityarajtiwari880-verisight-11ohzr?v=5ddb567da1a0cb2e2f12567300d7ff24)](https://m8ven.ai/mcp/adityarajtiwari880-verisight-11ohzr?s=readme)
38
40
  [![License](https://img.shields.io/badge/License-Apache_2.0-blue.svg)](LICENSE)
39
41
  [![Python Version](https://img.shields.io/badge/python-3.10%20%7C%203.11%20%7C%203.12-3776AB?logo=python&logoColor=white)](pyproject.toml)
@@ -43,6 +45,10 @@ Dynamic: license-file
43
45
  [![Next.js 16](https://img.shields.io/badge/Next.js-16.4.3-000000?logo=next.js&logoColor=white)](https://nextjs.org/)
44
46
  [![FastAPI](https://img.shields.io/badge/FastAPI-Production%20Ready-009688?logo=fastapi&logoColor=white)](https://fastapi.tiangolo.com/)
45
47
 
48
+ <p align="center">
49
+ <a href="https://mcprush.com/to/verisight-mcp?ref=badge"><img width="380" height="200" src="https://mcprush.com/badge/verisight-mcp/card.svg" alt="VeriSight MCP server" /></a>
50
+ </p>
51
+
46
52
  > **Sub-3ms Physical AI and multi-spectral visual forensics engine built on pure OpenCV 5 and SIMD acceleration.** Detects generative AI inpainting, deepfake manipulation, duplicate photo reuse, and presentation recapture attacks with zero heavyweight neural network dependencies.
47
53
 
48
54
  ---
@@ -77,6 +83,23 @@ Authentic Image ──> Generative AI Inpaint (Flux / Firefly) ──> Instant R
77
83
  | **Generalization & Drift** | Vulnerable to adversarial attacks & new diffusion models | **Grounded in mathematical physical optics & sensor physics** |
78
84
  | **Audit Explainability** | Black-box softmax probabilities | **Deterministic mathematical proof (ELA, PRNU, 2D FFT, Ghosts)** |
79
85
  | **Recapture Attack Defense**| None (Foiled by 2D screen or paper printout) | **Built-in Presentation Attack Detection (PAD)** |
86
+ | **External API Cost** | \$0.01–\$0.05/image via VLM/SaaS wrappers | **\$0.00 (100% offline/local C++ OpenCV 5)** |
87
+
88
+ ---
89
+
90
+ ## 🎯 Verified Real-World Evaluation Suite (`10/10` Passing in CI)
91
+
92
+ Verified automatically on every commit via [`backend/test_sanity.py`](backend/test_sanity.py) and [`backend/test_mcp_tools.py`](backend/test_mcp_tools.py):
93
+
94
+ | Test Scenario | Ground Truth | VeriSight Verdict | Risk Score |
95
+ | :--- | :--- | :--- | :---: |
96
+ | Genuine Unedited Photo | Authentic | `AUTO_APPROVE_CLAIM` | `10.9%` |
97
+ | Generative AI Inpainted Damage | Fraud | `REJECT_SYNTHETIC_TAMPER` | `99.4%` |
98
+ | Genuine Digital Coupon Screenshot | Authentic | `AUTO_APPROVE_CLAIM` | `30.8%` |
99
+ | AI Altered Coupon Code (`D` → `C`) | Fraud | `REJECT_SYNTHETIC_TAMPER` | `99.4%` |
100
+ | Printed Card Recapture Attack | PAD Attack | `REJECT_RECAPTURED_2D_MEDIA` | `97.5%` |
101
+ | Duplicate Photo Reuse (Sybil Ring) | Reuse Fraud | `REJECT_DUPLICATE_CLAIM_FRAUD` | `99.4%` |
102
+ | Selfie / Off-Subject in Produce or Invoice | Invalid Subject | `REJECT_INVALID_CLAIM_SUBJECT` | `Blocked` |
80
103
 
81
104
  ---
82
105
 
@@ -148,12 +171,17 @@ verisight server --port 8000
148
171
 
149
172
  VeriSight natively implements the **Anthropic Model Context Protocol (MCP)** specification (`protocol_version: 2024-11-05`). External LLM agents (**Claude Desktop**, **Cursor**, **Windsurf**, **LangChain**) can discover and call VeriSight tools autonomously via high-performance standard I/O (`stdio`).
150
173
 
151
- ### 1-Click Zero-Install (`uvx`)
174
+ ### 1-Click Zero-Install (`uvx` & `claude mcp`)
152
175
  Run directly via Astral `uvx` with zero manual cloning or environment setup:
153
176
  ```bash
154
177
  uvx verisight mcp
155
178
  ```
156
179
 
180
+ Or add directly to **Claude Code CLI**:
181
+ ```bash
182
+ claude mcp add --scope user verisight-mcp '--' uvx verisight mcp
183
+ ```
184
+
157
185
  ### Claude Desktop & Cursor Configuration
158
186
 
159
187
  Add VeriSight directly to your `claude_desktop_config.json` (or Cursor MCP settings):
File without changes
File without changes
File without changes