DITING

DITING is an LLM-driven autonomous penetration testing workbench designed for authorized security assessments.

DITING is an autonomous penetration testing workbench driven by large language models. It provides asset graphs, task exploration, multi-agent collaboration, traffic evidence, and manual approval workflows. Built on Go, Next.js, and PostgreSQL, it requires users to supply their own database and model configurations. It supports budget limits, asset scopes, evidence versioning, and traffic recording. The platform is intended strictly for authorized security targets and includes setup wizards and encrypted backup tools.

DITING at a glance

Best for: Software engineer
Runs on: Windows, Web, self-hosted
Pricing: Free
Released: October 2026
Category: AI coding tool
Website: github.com/bluechips-zhao/diting

Key features

Task and multi-agent coordination
Token and tool budget controls
Asset scope and interception management
Request and response evidence binding
Traffic recording and HAR export
Encrypted backup and recovery
Database initialization wizard

Use it for

Conducting autonomous penetration tests
Managing enterprise security assets
Recording and exporting network traffic
Configuring LLM agent budgets

Alternatives to DITING

Autonomous Pentest Agent — Dual-graph architecture for asset synchronization and multi-step attack planning.
ARTEX — Runs autonomous multi-agent penetration testing with a dual-graph architecture.
FakeLiang — Verifies attack paths step by step with traceable call records and evidence.
Artex Global — Autonomous multi-agent penetration testing with multilingual output support.
DeepTeam — It automates penetration testing using adversarial LLM simulations.
VulnClaw — It uses LLM agents to autonomously execute end-to-end security testing loops.
jevgrep — Reduces coding agent costs by providing targeted, relevant repository context.
CyberStrike — Autonomous AI agents execute recon, web, network, and cloud penetration testing.
DeepSeek Harness — Builds on a plugin architecture to organize documents, analyze tables, and write code.
Unlazy — Verifies AI agent output by running shell-based gate contracts against expectations.
PentestGPT — Runs autonomous multi-stage pipelines or interactive multi-LLM workflows for security tests.
Security Audit Skill — Automated multi-stage security auditing with independent verification.

All DITING alternatives