# Context Inspector

Bell Tuning™ for AI workflows. Reads the statistical bell curve of your context window — mean, σ, shape — so you tune the system on the signal, not the symptom. Catches context rot 3 steps before out…

## Quick Start

```bash
# Connect this server (installs CLI if needed)
npx -y @smithery/cli@latest mcp add kevinluddy39/contrarianai-context-inspector

# Browse available tools
npx -y @smithery/cli@latest tool list kevinluddy39/contrarianai-context-inspector

# Get full schema for a tool
npx -y @smithery/cli@latest tool get kevinluddy39/contrarianai-context-inspector analyze_context

# Call a tool
npx -y @smithery/cli@latest tool call kevinluddy39/contrarianai-context-inspector analyze_context '{}'
```

## Direct MCP Connection

Endpoint: `https://contrarianai-context-inspector--kevinluddy39.run.tools`

## Tools (4)

- `analyze_context` — Full context analysis — domain/user alignment, statistics, bell curve data, per-chunk breakdown.
- `get_bell_curve` — Quick bell curve summary (mean, σ, histogram) for domain or user alignment of a text.
- `get_chunks` — Per-chunk alignment scores with top-N highest and lowest-scoring chunks.
- `compare_alignment` — Side-by-side domain vs user alignment comparison.

```bash
# Get full input/output schema for a tool
npx -y @smithery/cli@latest tool get kevinluddy39/contrarianai-context-inspector <tool-name>
```

---

License: MIT