CyberStrikeAI is a Go-based security automation platform that orchestrates agents, tools, and knowledge for authorized security operations. It provides AI channel configuration, multi-agent orchestration, and an auditable workflow with CI-like governance.
Collecting history — the radar snapshots this repo daily. The trend line appears after 3 days of data (1 so far).
What it is
CyberStrikeAI is described as a system that connects planning, execution, human oversight, evidence, and replay in one auditable workspace. It is built in Go and combines Eino-powered agents, MCP-native tools, RAG knowledge, visual workflows, and attack-chain modeling for authorized security operations.
How it works
Key capabilities include agentic execution that translates intent into actions, Eino orchestration with single-agent and multi-agent modes, graph workflows that combine agents, tools, conditions, and outputs, and role-based testing for security scenarios. It includes tool categories across scanning, vulnerability, subdomain, API security, container and cloud security, binary analysis, exploitation, password cracking, forensics, CTF utilities, and system helpers. Governance features cover human-in-the-loop approvals, RBAC, and auditability. Built-in C2 and WebShell management are included for authorized high-risk operations.
Getting started
- Quick Start: One-Command Deployment via run.sh after cloning and using Go/Python prerequisites. The README shows how to clone, cd, chmod +x run.sh, and run ./run.sh. The run.sh script is documented to set up environments, install dependencies, build, and start the server.
- First-time configuration includes AI Channel Configuration in System Settings to add or edit channels with provider, base_url, api_key, model, and token limits. YAML sample is provided for default channel configuration. Direct config.yaml edits are supported.
- Alternative launch methods include direct Go run or manual build commands with --https or server flags.
Commands and snippets exactly as written in the README are:
git clone https://github.com/Ed1s0nZ/CyberStrikeAI.git
cd CyberStrikeAI
chmod +x run.sh && ./run.sh
and
# Direct Go run (set up env yourself); add --https to match run.sh defaults
go run cmd/server/main.go --https
# Manual build
go build -o cyberstrike-ai cmd/server/main.go
./cyberstrike-ai --https
Networking defaults note the use of --https and config.yaml with local self-signed TLS, with an option to use --http for plain HTTP.
Recent releases
Latest release is v1.7.11 (2026-07-28) with a finalization protocol and explicit evidence policy to enforce structured finalization and avoid intermediate outputs. Other recent entries include v1.7.10 (2026-07-28), v1.7.9 (2026-07-23), v1.7.8 (2026-07-23) adding AI channel configuration capabilities and per-conversation role synchronization. The changelog text for v1.7.10 and earlier is summarized in the notes listed in the README.
Traction
Stars: 5541. Forks: 879. Open issues: 41. Language: Go. License: Apache-2.0. Created: 2025-11-08. Last push: 2026-08-04.
Behind the repo
Not provided in the excerpt. No startup/company linkage is stated beyond the repository description and documentation sections.
Caveats
- License: Apache-2.0 is stated.
- Security advisory notes emphasize authorized testing and reviewing security model and hardening guides before enabling high-risk tools.
- Prerequisites include Go 1.25+ and Python 3.10+ as shown in Quick Start.
- Configuration guidance exists for server settings, TLS, and AI channel configuration, including sample YAML and references to configuration files.