v0.1.0 -- Apache-2.0

One CLI for AI Agent Security

Credential detection with scope drift analysis, config integrity signing, runtime monitoring, and supply chain verification. Works with Node.js, Python, Go, and MCP server projects.

$ npx opena2a init

Instant Security Assessment

Run opena2a init in any project to get a trust score, credential findings, scope drift alerts, and prioritized next steps.

Terminal
$ opena2a init --verbose
OpenA2A Security Initialization v0.1.0
Project acme-agent v2.1.0
Type Node.js + MCP server
Security Posture
-----------------------------------------------
Credential scan 4 findings
.gitignore present
.env protection NOT in .gitignore
Lock file package-lock.json
MCP config found
-----------------------------------------------
Trust Score 15 / 100 [Grade: F]
Scope Drift Detected
-----------------------------------------------
DRIFT-001 Google Maps key may access Gemini AI
DRIFT-002 AWS key may access Bedrock AI

Core Commands

Six built-in commands plus 11 adapter-backed tools from the OpenA2A ecosystem.

opena2a init

Assess your project's security posture

Detects project type, scans for credentials, checks hygiene, calculates trust score (0-100), and provides prioritized next steps.

opena2a protect

Detect and migrate hardcoded credentials

Finds API keys across all source files. Language-aware replacements for JS, Python, Go, Ruby, Java, and Rust. Stores secrets in vault or .env.

opena2a guard

Config file integrity protection

Signs config files with SHA-256 hashes. Detects tampering, tracks unsigned files, maintains a signature store.

opena2a runtime

Agent runtime monitoring

Wraps ARP (Agent Runtime Protection) to monitor process, network, and filesystem activity. Protocol-aware for MCP and A2A.

opena2a verify

Binary integrity verification

Compares installed package hashes against the Trust Registry. Detects supply chain tampering across all OpenA2A tools.

opena2a scan

150+ security checks via HackMyAgent

Adapter-backed scanning with attack simulation, auto-fix suggestions, and compliance reporting.

New in v0.1.0

Scope Drift Detection

API keys provisioned for one service often silently grant access to others. A Google Maps key can call Gemini. An AWS S3 key may reach Bedrock. OpenA2A detects these cross-service privilege escalations.

DRIFT-001Google Cloud

Google API Key Gemini Drift

A key provisioned for Google Maps silently grants access to Gemini AI models. Attackers can run LLM workloads billed to your account.

Unintended access: Gemini API access
DRIFT-002AWS

AWS Access Key Bedrock Drift

An AWS key intended for S3 or EC2 may also grant access to Bedrock LLM models due to over-provisioned IAM policies.

Unintended access: Bedrock API access

Detected Credential Patterns

Regex-based detection with language-aware replacements for JS, Python, Go, Ruby, Java, and Rust.

IDPatternPrefixSeverity
CRED-001Anthropic API Keysk-ant-api*Critical
CRED-002OpenAI API Keysk-*, sk-proj-*Critical
CRED-003GitHub Tokenghp_*, ghs_*High
DRIFT-001Google API KeyAIza*High
DRIFT-002AWS Access KeyAKIA*High
CRED-004Generic API Keyapi_key=, secret_key=Medium

Smart Command Discovery

Four input modes so you always find the right command, whether you know exactly what you need or just have a vague idea.

Interactive Wizard

Run with no arguments for a guided menu.

$ opena2a

Semantic Search

Search by concept with domain expansion.

$ opena2a ~google drift

Context-Aware Suggestions

Get recommendations based on your project state.

$ opena2a ?

Natural Language

Describe what you want in plain English.

$ opena2a "find leaked credentials"

CI/CD Integration

All commands support --format json and --ci for pipeline integration.

.github/workflows/security.yml
- name: Security assessment
  run: npx opena2a init --ci --format json > report.json

- name: Credential check
  run: |
    npx opena2a protect --dry-run --ci --format json > creds.json
    jq -e '.totalFound == 0' creds.json

- name: Config integrity
  run: npx opena2a guard verify --ci

Unified Security Ecosystem

The CLI orchestrates 11 specialized security tools through adapter commands. Each installs on first use -- no manual setup required.

scan
150+ security checks
secrets
Credential management
benchmark
222 attack scenarios
runtime
Runtime monitoring
research
HMA Researcher
Autonomous research
hunt
HMA Hunter
Attack decomposition
train
Training environment
crypto
CryptoServe
PQC readiness
identity
Agent identity
broker
Credential broker
dlp
Data loss prevention
guard
ConfigGuard
Config integrity

Get Started in 30 Seconds

No signup, no configuration, no dependencies. One command to assess your project.

$ npx opena2a init # assess
$ npx opena2a protect # fix
$ npx opena2a guard sign # harden