8 RESULTS · “instructions”
v0.1.6 · MIT
Fork of @deepseek-ai/dsh-agent-instructions that delivers the workspace instruction baseline as system-prompt sections, keeping the upstream mid-session reminder path.
v1.1.78 · Apache-2.0
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
v1.1.71 · MIT
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。
v1.1.72-dsh021rc2.4 · MIT
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）
v0.28.0 · MIT
Linter for AI agent context files and MCP configs: CLAUDE.md, AGENTS.md, .cursorrules, .mcp.json - catches broken paths, wrong commands, leaked secrets
v1.3.1 · MIT
AI-assisted toolkit for making HTML5 games with LittleJS: the engine, templates, and agent instructions that work with Claude Code, Codex, Cursor and Copilot, so the AI builds your game instead of rebuilding an engine. Installable as a Claude Code plugin.
v1.4.9 · no license
Don't let AI destroy your hard work! HardStop is a rock-solid protection for AI-generated commands. Pre-execution safety validation for Claude Code, Claude Cowork. Catches dangerous commands before they run: whether from AI mistakes, hallucinations, prompt injection, or misunderstood instructions. Seatbelts for the agentic AI era.