Spaces:

ZeeshanSaud
/

CodeTruthAgent-V3

Running

File size: 5,838 Bytes

<!DOCTYPE html>
<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>