
Dea
Evidence from strace, py-spy, perf, lsof and tcpdump for agents debugging live processes
- Pricing
- Open source
- Platforms
- CLI, Linux, macOS
- Runs
- Locally
- Version
- 0.4.0
- Listing updated
- 2 days ago
About Dea
Dea is an open-source MCP server for developer tools. It runs locally on your machine and is started by your AI client.
Rather than guessing from source code, the agent attaches to a running process or container and each reading is numbered and logged with its time, the tool that took it and the raw output on your disk. The site shows a leaky Node.js server where successive memory snapshots show an array growing from 2.0 KB to 87.7 MB, flagged as a HIGH, observed finding.
Commands cover memory, CPU profiles, thread stacks, sockets, file descriptors, syscalls, network capture and containers across Node, Python, Go, Java and native processes. It is read-only: it never kills, patches or sets breakpoints, never runs sudo for you, and asks before packet capture or opening a Node inspector. Common token formats are redacted from what the agent sees.
The code is MIT-licensed and, per the site, sends no telemetry and has no update checks.
Install
Run in a terminal
claude mcp add dea -- npx -y dea-agentsAdd to claude_desktop_config.json
{
"mcpServers": {
"dea": {
"command": "npx",
"args": [
"-y",
"dea-agents"
]
}
}
}Add to ~/.cursor/mcp.json
{
"mcpServers": {
"dea": {
"command": "npx",
"args": [
"-y",
"dea-agents"
]
}
}
}Add to .vscode/mcp.json
{
"servers": {
"dea": {
"type": "stdio",
"command": "npx",
"args": [
"-y",
"dea-agents"
]
}
}
}Add to ~/.codeium/windsurf/mcp_config.json
{
"mcpServers": {
"dea": {
"command": "npx",
"args": [
"-y",
"dea-agents"
]
}
}
}Add to your client's MCP settings (most use this shape)
{
"mcpServers": {
"dea": {
"command": "npx",
"args": [
"-y",
"dea-agents"
]
}
}
}Tools this server provides (2)
Taken from the project documentation; the running server may differ.
- snapshot_memory
- Takes a heap or memory snapshot of the attached process.
- diff_runs
- Compares readings between runs to show growth.
What it does
- Numbered evidence
- Every finding cites evidence ids and is marked observed or inferred.
- Wide process coverage
- Memory, CPU, threads, sockets, files, syscalls and containers.
- Read-only by design
- No kills, patches or breakpoints.
- Asks first
- Packet capture and Node inspector need an explicit yes.
- Python and CLI use
- Also callable from LangChain, LlamaIndex or CrewAI agents.
Good for
- Find why a server's memory keeps growing
- See who holds a port or leaks descriptors
- Diagnose OOM-killed containers
- Profile where a slow Python process spends time
Pricing
Open source
MIT licence
Prices change. Check debuganything.dev before you buy.
Dea FAQ
Does it need root?
Most tools do not; when one does it prints the exact command for you.
Does anything leave my machine?
The site says no.
Is it open source?
Yes, MIT.
Is Dea free?
Yes. Dea is open source, so you can use and self-host it for free.
How do I add Dea to Claude Code?
Run this command in your terminal: claude mcp add dea -- npx -y dea-agents
What are the best alternatives to Dea?
The closest alternatives to Dea are REA, Codebase Memory, GitHub, Serena MCP: the IDE for your agent and XERJ.
Alternatives to Dea

REA
MCP serverCLI and MCP server that lets coding agents reverse engineer apps and binaries
Runs locally- Open source
- 464K GitHub stars
- Developer tools

Codebase Memory
MCP serverLocal MCP server that builds a code knowledge graph for AI coding agents
Runs locally- Open source
- 48.7K GitHub stars
- Knowledge and memory

GitHub
MCP serverOfficial MCP server connecting AI assistants to GitHub repos, issues and workflows
Not checked yet- Remote
- Open source
- 34.5K GitHub stars
- Developer tools

Serena MCP: the IDE for your agent
MCP serverMCP toolkit giving coding agents symbol-level code retrieval and editing
Runs locally- Open source
- 31.5K GitHub stars
- Developer tools

XERJ
MCP serverLocal search engine for AI agents with cited answers over your code and docs
Runs locally- Open source
- 19.2K GitHub stars
- Knowledge and memory

Ouroboros
MCP serverSpec-first layer that pins requirements and verifies results of AI coding agent work
Runs locally- Open source
- 6.4K GitHub stars
- Developer tools