Agentic Store
A local toolkit for code work: lint and secret-scan files on disk, audit dependencies against the CVE database, and read or change a GitHub repo — with every write shown as a preview first.Read full description
agenticstoreLABSCO SUMMARY
Correlating several weak signals into one finding is the part a firmware scan leaves undone.
EMBA produces a great deal of output, and the individually-unremarkable items are where the real problem usually hides — a SUID binary is fine, a weak credential is fine, the two together are not. An engine that combines signals and then explains the path between them turns a report you skim into a report you act on. Working from stored output rather than re-scanning also means the analysis is cheap to repeat as your questions change.
What it is
A reader for EMBA's firmware analysis output. It parses results you already have rather than running the scan again, then normalises and reasons over them.
What you get
- Parsing EMBA results across kernel, services, credentials, crypto, the software bill of materials, binaries and more
- Filesystem-aware analysis covering SUID binaries, secrets, weak crypto and exposed services
- A correlation engine that combines several signals into one high-risk finding
- An engine that explains the attack path behind a finding
Requirements
Linux with Python 3.10 or newer and EMBA itself. It works against existing output, so no rescan is required.
Setup effort
Build from source — git clone https://github.com/0xbuz3R/EMBA-MCP.git
USE CASES
01
Making sense of a scan you already ran
Parse and normalise the existing output.
02
Finding what actually matters
Multi-signal correlation rather than a flat finding list.
03
Explaining the risk
The attack path behind a high-risk finding.
BEFORE YOU INSTALL
What to know first
Requires
Python 3.10+
Where it runs
The documented client runs the server on your machine.
INSTALLATION
Connect your client
This one is built from source rather than installed with a single command.
Server setup
TERMINAL
git clone https://github.com/0xbuz3R/EMBA-MCP.git
cd EMBA-MCP
python -m venv venv
source venv/bin/activate (macOS/Linux) or .\venv\Scripts\activate (Windows)
pip install -e .Register the server in ~/.config/claude/claude_config_desktop.json using the mcpServers block below, then fully restart Claude Desktop
MCP Clients
Claude Desktop
CLAUDE DESKTOP · JSON
{
"mcpServers": {
"emba": {
"command": "<YOUR_PATH_TO_PYTHON>",
"args": [
"-m",
"emba_mcp.mcp_server"
],
"env": {
"EMBA_HOME": "<YOUR_EMBA_HOME>"
}
}
}
}
