Spaces:
Running
Running
| <html lang="en"> | |
| <head> | |
| <meta charset="UTF-8"> | |
| <meta name="viewport" content="width=device-width, initial-scale=1.0"> | |
| <title>CodeTruth Agent V3</title> | |
| <style> | |
| body { font-family: monospace; max-width: 860px; margin: 40px auto; padding: 0 20px; background: #0d1117; color: #e6edf3; line-height: 1.7; } | |
| h1 { color: #58a6ff; border-bottom: 1px solid #30363d; padding-bottom: 10px; } | |
| h2 { color: #79c0ff; margin-top: 32px; } | |
| h3 { color: #a5d6ff; } | |
| .badge { display: inline-block; background: #1f6feb; color: white; padding: 2px 10px; border-radius: 4px; font-size: 12px; margin: 2px; text-decoration: none; } | |
| .badge.green { background: #238636; } | |
| .badge.orange { background: #9e6a03; } | |
| pre { background: #161b22; border: 1px solid #30363d; border-radius: 6px; padding: 16px; overflow-x: auto; font-size: 13px; } | |
| table { border-collapse: collapse; width: 100%; } | |
| th, td { border: 1px solid #30363d; padding: 8px 12px; text-align: left; } | |
| th { background: #161b22; color: #79c0ff; } | |
| a { color: #58a6ff; } | |
| .philosophy { background: #161b22; border-left: 4px solid #58a6ff; padding: 12px 16px; margin: 20px 0; font-style: italic; } | |
| .status { color: #3fb950; } | |
| </style> | |
| </head> | |
| <body> | |
| <h1>CodeTruth Agent V3</h1> | |
| <div> | |
| <a href="https://github.com/Zeeshan78699/CodeTruthAgent" class="badge">GitHub</a> | |
| <a href="https://doi.org/10.5281/zenodo.20669542" class="badge green">DOI M1 Core</a> | |
| <a href="https://doi.org/10.5281/zenodo.21005444" class="badge green">DOI M1 Extension</a> | |
| <a href="https://doi.org/10.5281/zenodo.20706591" class="badge green">DOI M2 Core</a> | |
| <a href="https://doi.org/10.5281/zenodo.21005972" class="badge green">DOI M2 Deep Resolution</a> | |
| <span class="badge orange">GPL-3.0</span> | |
| </div> | |
| <div class="philosophy"> | |
| AI imagines. CodeTruth checks. Nature tests. Humans decide. | |
| </div> | |
| <h2>What It Does</h2> | |
| <p>Before an AI agent modifies a codebase, it must answer two questions deterministically:</p> | |
| <pre> | |
| Question 1: What is this repository? ← Module 1 | |
| Question 2: What calls what inside it? ← Module 2 | |
| </pre> | |
| <p>No AI models. No guessing. Pure static analysis with governed, evidence-based decisions.</p> | |
| <h2>Pipeline</h2> | |
| <pre> | |
| Repository | |
| ↓ | |
| Module 1 — Repository Cognition Engine | |
| Domain: ENERGY_SYSTEM / AEROSPACE / FINANCE / MEDICAL ... | |
| Framework: pandapower / pyNastran / CCXT / PyDICOM | |
| Gate: APPROVED / REVIEW_REQUIRED / BLOCKED | |
| ↓ (if APPROVED) | |
| Module 2 — Repository Graph Intelligence | |
| Call graph: 1,916,272 calls resolved | |
| Deep Resolution: 394,796 additional (+25.9%) | |
| Language adapters: Python / SQL / C# / Go | |
| ↓ | |
| Module 3 — Repository Reasoning Engine (in progress) | |
| Data flow tracing / Return type inference | |
| </pre> | |
| <h2>Validation — Module 1</h2> | |
| <table> | |
| <tr><th>Metric</th><th>Result</th></tr> | |
| <tr><td>Repositories validated</td><td>32/32 PASS + 11/11 cross-domain</td></tr> | |
| <tr><td>Engineering domains</td><td>16 specific domains</td></tr> | |
| <tr><td>File extensions</td><td>218 across 25 domain blocks</td></tr> | |
| <tr><td>Crashes</td><td>0</td></tr> | |
| <tr><td>Governance gate</td><td>APPROVED / REVIEW_REQUIRED / BLOCKED</td></tr> | |
| </table> | |
| <h2>Validation — Module 2</h2> | |
| <table> | |
| <tr><th>Metric</th><th>Result</th></tr> | |
| <tr><td>Repositories validated</td><td>76/76 PASS</td></tr> | |
| <tr><td>Files processed</td><td>54,435</td></tr> | |
| <tr><td>Baseline calls resolved</td><td>1,521,476</td></tr> | |
| <tr><td>Deep Resolution additional</td><td>394,796 (+25.9%)</td></tr> | |
| <tr><td>Total calls resolved</td><td>1,916,272</td></tr> | |
| <tr><td>Language adapters</td><td>Python, Oracle SQL, C#, Go</td></tr> | |
| <tr><td>DR resolvers proven</td><td>7</td></tr> | |
| <tr><td>Crashes</td><td>0</td></tr> | |
| </table> | |
| <h2>11-Domain Cross-Validation</h2> | |
| <pre> | |
| fastapi API_SERVICE FastAPI ✅ | |
| pytorch ML_PIPELINE PyTorch ✅ | |
| django WEB_APPLICATION Django ✅ | |
| biopython SCIENTIFIC_COMPUTING BioPython ✅ | |
| OpenMDAO AEROSPACE_STRUCTURAL_SIMULATION OpenMDAO ✅ | |
| lasio WELL_LOGGING Lasio ✅ | |
| welleng DRILLING_SYSTEM welleng ✅ | |
| ccxt FINANCE_SYSTEM CCXT ✅ | |
| pandapower ENERGY_SYSTEM pandapower ✅ | |
| poliastro SPACE_SYSTEM poliastro ✅ | |
| pydicom MEDICAL_SYSTEM PyDICOM ✅ | |
| </pre> | |
| <h2>How to Run</h2> | |
| <pre> | |
| git clone https://github.com/Zeeshan78699/CodeTruthAgent | |
| cd CodeTruthAgent | |
| python -m venv .venv | |
| .venv\Scripts\activate # Windows | |
| pip install -r requirements.txt | |
| # Module 1 only | |
| python run_m1.py "C:\repos\your_repo" | |
| # Module 2 only | |
| python run_m2.py "C:\repos\your_repo" | |
| # Full pipeline | |
| python pipeline.py "C:\repos\your_repo" | |
| </pre> | |
| <h2>DOIs</h2> | |
| <table> | |
| <tr><th>Record</th><th>DOI</th></tr> | |
| <tr><td>Module 1 Core</td><td><a href="https://doi.org/10.5281/zenodo.20669542">10.5281/zenodo.20669542</a></td></tr> | |
| <tr><td>Module 1 Extension Layer</td><td><a href="https://doi.org/10.5281/zenodo.21005444">10.5281/zenodo.21005444</a></td></tr> | |
| <tr><td>Module 2 Core</td><td><a href="https://doi.org/10.5281/zenodo.20706591">10.5281/zenodo.20706591</a></td></tr> | |
| <tr><td>Module 2 Deep Resolution</td><td><a href="https://doi.org/10.5281/zenodo.21005972">10.5281/zenodo.21005972</a></td></tr> | |
| </table> | |
| <h2>Status</h2> | |
| <pre> | |
| Module 1 — Repository Cognition ✅ COMPLETE FROZEN | |
| Module 2 — Repository Graph Intel ✅ COMPLETE FROZEN | |
| Module 3 — Repository Reasoning 🔄 IN PROGRESS | |
| Modules 4-6 📋 PLANNED | |
| </pre> | |
| <p style="margin-top:40px; color:#6e7681; font-size:12px;"> | |
| CodeTruth Agent V3 — <a href="https://github.com/Zeeshan78699/CodeTruthAgent">github.com/Zeeshan78699/CodeTruthAgent</a> | |
| </p> | |
| </body> | |
| </html> |