Bagel

Dev workstation security scanner from BoostSecurity. Checks Git config, SSH keys, npm settings, shell history, cloud credentials, IDE plugins, GitHub CLI, and AI tool configs for security gaps. Privacy-first — only metadata leaves the machine.

I’m a contributor to this project.

Source: boostsecurityio/bagel

What It Does

Bagel scans your local dev environment for misconfigurations and exposed secrets that attackers target in supply chain compromises. It runs 9 probes and 8 secret detectors across common developer tools.

bagel scan  -->  9 probes + 8 secret detectors  -->  findings report

No data leaves your machine. Bagel reads local configs and reports findings locally. In --report mode, only metadata (finding type, severity) is sent — never file contents, secrets, or paths.

Install

brew install boostsecurityio/tap/bagel

Or from source:

git clone https://github.com/boostsecurityio/bagel.git
cd bagel && go build -o bagel .

Requires Go 1.25+. License: GPL-3.0.

Usage

bagel scan              # scan with default settings
bagel scan --strict     # CI/CD mode -- exit 1 on any finding
bagel scan --report     # send metadata to BoostSecurity dashboard
bagel scrub             # interactive cleanup of detected secrets

Probes (9)

Probe	What it checks
Git	Commit signing, credential helpers, safe directory settings
SSH	Key types, permissions, agent forwarding config
npm	Registry config, auth tokens in .npmrc, publish settings
Environment Variables	Secrets in shell rc files (.bashrc, .zshrc, .profile)
Shell History	Credentials and tokens leaked in command history
Cloud Credentials	AWS, GCP, Azure credential files and their permissions
JetBrains	IDE plugin security, stored credentials, project settings
GitHub CLI	Token scopes, auth config, stored credentials
AI Tools	API keys and configs for Copilot, Claude, OpenAI, etc.

Secret Detectors (8)

Scans common locations for leaked credentials:

• AWS access keys and secret keys
• GCP service account JSON keys
• Azure client secrets
• GitHub personal access tokens
• npm auth tokens
• SSH private keys (checks permissions too)
• Generic API keys in env files
• AI tool API keys (OpenAI, Anthropic, etc.)

CI/CD Integration

Use --strict to fail pipelines when findings exist:

# GitHub Actions
- name: Dev workstation audit
  run: bagel scan --strict

This catches developers committing with misconfigured Git settings, expired SSH keys, or leaked credentials in rc files.

Scrub Command

bagel scrub provides interactive cleanup:

1. Shows each finding with context
2. Offers to fix automatically where possible (permissions, config changes)
3. For secrets: shows location, offers to rotate/remove

Why This Matters

Supply chain attacks increasingly target developer machines — not just CI/CD. A compromised .npmrc with an auth token, an AWS key in .bash_history, or a GitHub PAT with repo scope in an unsecured config file are all realistic attack vectors. Bagel catches these before an attacker does.