0sec-labs/0

The AI offensive security OS - open, self-improving, extensible. Find, verify, and fix vulnerabilities 24/7.

0 – Open‑source security workflows powered by LLMs

What it is – 0 (pronounced “zero”) is a research‑preview tool that lets you run AI‑assisted security analyses from a browser UI, a CLI, or via the MCP (Multi‑Client Protocol) that other agents can talk to. It ships with a set of ready‑made templates for common security tasks (code‑review, dependency scanning, API testing, smart‑contract audit, etc.) and records the run results, evidence, and remediation suggestions.

How it works

  • You install a small binary (curl … | bash) that adds 0 to ~/.0/bin.
  • Configure a model provider (any LLM that supports the LiteLLM API) once; the same model is used for all workflows.
  • Pick a template (0 workflow list --templates) and run it against a target directory or repo (0 workflow run …).
  • Results are stored locally and can be inspected in the web UI (0 web) or via the CLI (0 runs list).
  • The engine can be exposed as an MCP server so other tools (IDE extensions, CI pipelines, custom agents) can create, list, and control workflows programmatically.

Key capabilities

Area Built‑in support
Assess Repo, dependency, API, web‑config, penetration‑test scopes, package behaviour, smart contracts, native binaries
Verify Replay evidence for findings
Fix Generate candidate patches with regression‑test scaffolding
Research Deep source investigation, custom investigative steps
Learning Keeps a local knowledge base of runs, notes, and improvement history
Integrations Plugins for GitHub, Elastic, Semgrep, Snyk, Linear, Jira, Cloudflare, Slack, Teams; also usable from GitHub Actions
MCP Remote or local engine access, session sharing, token‑based auth

Typical use cases

  • Security teams wanting a quick, AI‑augmented audit of a new codebase without wiring up a full CI pipeline.
  • Developers who need on‑demand suggestions for fixing high‑impact vulnerabilities discovered by the tool.
  • Researchers prototyping new security‑analysis steps and sharing them as reusable templates.
  • Automation scripts that call 0 mcp-server to run scheduled scans across many repositories.

Getting started

# Install (macOS Apple Silicon or Linux x64/ARM64)
curl -fsSL https://raw.githubusercontent.com/0sec-labs/0/main/install.sh | bash
export PATH="$HOME/.0/bin:$PATH"

# Open the web UI
0 web

# Run a template against a repo
0 workflow run --template security-research \
    --target /path/to/repo --workspace /path/to/repo \
    --format json > run.json

Development & contribution – The project is built with Bun, React, and OpenTUI. After cloning, run npm run dev (or bun run dev) to start a local dev server; contribution guidelines are in .github/CONTRIBUTING.md.

Status & safety – Marked research preview. The interactive terminal UI is deprecated, but the headless CLI, automation hooks, and MCP remain functional. Users are encouraged to run only on authorized systems and can enable the optional scope plugin to enforce target boundaries.

License – Dual‑licensed under MIT OR Apache‑2.0.

All information above is taken directly from the repository’s README; no additional features have been inferred.

Related

  • Project
    mlflow/mlflow

    MLflow is an Apache‑2.0‑licensed open‑source platform for managing the full lifecycle of machine‑learning and generative‑AI projects. It provides experiment tracking, model registry, deployment tools, and a new GenAI stack (tracing, evaluation, prompt registry, optimization, and an AI gateway) with integrations for dozens of LLM providers and agent frameworks.

  • Project
    Kritt-ai/open-kritt

    open·kritt is an open‑source, self‑hosted platform that orchestrates AI agents (via Codex, Claude Code, Grok Build, etc.) to run custom, prompt‑driven security workflows on code repositories. It builds focused tasks, runs them in parallel Docker containers, validates findings, de‑duplicates and ranks them, and exports a complete report bundle. Designed for bug‑bounty researchers and DevSecOps, it offers a web UI and a headless CLI, supports any model provider, and is released under AGPL‑3.0.

  • Project
    awslabs/aidlc-workflows

    AI-DLC is a structured, harness-neutral framework that turns AI coding assistants into verifiable, audited software delivery workflows with human approval gates and multi-stage lifecycle management.

  • Project
    bytedance/deer-flow

    DeerFlow is an open‑source framework that lets you build autonomous AI agents capable of orchestrating sub‑agents, using tools (web search, sandboxed code execution, RAG), and managing memory. It supports many LLM providers, offers a web UI, a terminal UI, and integrates tracing services, all deployable via Docker.

  • Project
    agent0ai/agent-zero

    Agent Zero is an open‑source Docker‑based framework that gives an LLM‑driven AI agent a full Linux desktop, a browser with DOM‑annotation, live document editors, and a plugin ecosystem. Agents can run GUI apps (e.g., Blender, LibreOffice), edit files, browse and annotate web pages, and even spawn sub‑agents, all while you watch and intervene through a web UI. Installation is via a cross‑platform launcher or a one‑line script, and the system supports project‑scoped workspaces, host‑machine bridging, time‑travel snapshots, and extensive community plugins.