DeepSeek Harness 破甲：让所有模型都能破甲，不同模型可换不同提示词；默认提示词面向国模「小码酱」。Jailbreak for every model — swap prompts per model. 求 Star 收藏 ⭐ 如要二开，请查看署名要求部分

Local security audit for AI API relays and LLM proxies: detects prompt injection, model substitution, tool-call rewriting, SSE anomalies, error leakage, and Web3 wallet risks. Website: https://apirelayaudit.com

Read-only security & compliance plugin for DeepSeek Harness: prompt-injection detection, Chinese-PII redaction, and local configuration audit with redacted, reproducible reports.

Prompt-injection, jailbreak, and secret-leak detection with allow/ask/block interception for DeepSeek Harness: an Aho-Corasick pattern engine and heuristics ported from the Prompt-Injection-Payloads, Jailbreak-Detector, and Secret-Key-Leaker-Detect assets

Open-source antivirus for AI agents: block risky tools, secret access, prompt injection, malicious packages, MCP servers, plugins, and skills at runtime.

DeepSeek Harness 破甲 (mirror of YuJunZhiXue/dsh-purge) — 上游即将删库，此为本仓库全量镜像：含全部分支/73 个 tag/完整提交历史。原作者署名与二开要求见 README 署名章节。

dsh-purge 社区适配版：上游已停更，这里让它继续跑在 DeepSeek Harness 0.2.1-rc.2 上（基于 1.1.72）

The vendor-neutral protocol for AI agent safety & security — and the neutral benchmark that ranks the vendors.