安全与治理
插件
Claude Code-style declarative permission rules for DeepSeek Harness: ordered allow/deny/ask rules with tool-name, argument (glob/regex), and workspace-path matching on the tools/pre-execute waterfall, session-log audit, and HMR reload.
ai-agent
ai-agents
ai-safety
allow-deny-ask
安全与治理
技能
Custom-built for dsh (DeepSeek Harness): persistent Ghidra headless daemon + reverse-engineering agent skills (triage/unpack/decompile/pwn/pcap/APK) with circuit breakers against agent failure loops. 为 dsh 深度定制:会熔断、能从复盘进化的逆向 skill 家族。
agent-skills
ai-agent
android-reverse-engineering
binary-analysis
安全与治理
插件
Authorized security-assessment profile for DeepSeek Harness: scoped network tools, container-backed shell, authorization guard, durable evidence, 21 security skills, 7 specialist subagents
deepseek-harness
dsh-plugin
penetration-testing
security-assessment
安全与治理
插件
Inline HTML + KaTeX renderer for DeepSeek Harness (DSH) chats - sandboxed iframes, theme-adaptive, height-accurate, zero network
deepseek
dsh
dsh-plugin
html-renderer