dedroom 0.1.0

Unified agent runtime: loop detection + context compression for AI coding agents

DedrooM

Loop detection + context compression for AI coding agents.

DedrooM sits between your AI agent (Claude Code, Codex, Aider, Cursor, Cline, OpenCode) and the LLM provider to:

• Detect and block infinite loops — saves wasted API calls when tools repeat
• Compress context — reduces token usage by 60–95% without changing behavior
• Redact sensitive data — strip API keys, tokens, and secrets from tool outputs
• Track ROI — attribution engine shows exactly how much each tool saves

Quick Start

pip install dedroom

# Wrap an AI agent through the proxy
dedroom wrap claude         # Claude Code
dedroom wrap codex           # OpenAI Codex CLI
dedroom wrap aider           # Aider
dedroom wrap cursor          # Cursor Editor
dedroom wrap cline           # Cline (VS Code)
dedroom wrap opencode        # OpenCode

Commands

Command	Description
dedroom wrap <agent>	Start proxy + launch agent through it
dedroom unwrap <agent>	Restore agent config to pre-wrap state
dedroom doctor	Run diagnostics (proxy, routing, savings)
dedroom proxy	Start standalone proxy server
dedroom dash	Launch the TUI dashboard

Python API

from dedroom import DedrooM

pipeline = DedrooM("""
loop_detection:
  max_repeats: 3
""")

# Check for loops
verdict = pipeline.verify("write_file", '{"path": "/tmp/x.txt"}')
# 0 = Allow, 1 = Warn, 2 = BlockRetry, 3 = BlockHalt

# Compress content
compressed = pipeline.compress(tool_output, content_type="code")

# Full pipeline
result = pipeline.process_tool("write_file", '{}', tool_result)
print(f"Blocked: {result['is_blocked']}")
print(f"Saved {result['original_tokens'] - result['compressed_tokens']} tokens")

Development

# Clone and build
git clone https://github.com/Devaretanmay/dedroom
cd dedroom
cargo build -p dedroom-cli -p dedroom-proxy

# Install Python package in dev mode
pip install -e .

# Run tests
pytest python/tests/