209 results
Compliance skill hub provider for DeepSeek Harness: install cross-border compliance skills (CCPA/GDPR/COPPA/HIPAA x check/guard/compliance, 9 skills) from a line-level JSON catalog via ctx.skills (list/get), with DISCLOSURE v0.2 open-data-layer disclosure.
SQL 优化模式：EXPLAIN 分析、索引策略、N+1 解决、查询改写。受 wshobson/agents（38k★ MIT）启发。
Plugin Cordis para o DeepSeek Harness v0.1: interceta a camada HTTP (auth barrier sobre o fallback da Web UI) e orquestra subprocessos de longa duracao com ciclo de vida reversivel.
Agent harness 配置系统：agents/skills/hooks/rules 四层分工、instincts 持续学习（自动学模式、聚簇成技能）、AgentShield 安全审计（secrets/权限/hook 注入/MCP 风险）、Unified Memory Vault 跨 harness 共享、GateGuard 破坏性命令门卫。受 affaan-m/ECC（185k★ MIT）启发。
One install for SpecFlow, GitFlow, Guardian, and Code Intel on DeepSeek Harness.
Fail-closed secret-like payload guard for DeepSeek Harness tool calls.
Data-loss-prevention plugin for DeepSeek Harness: a non-configurable tool guard floor, tool-result redaction, and fail-closed telemetry redaction
Egress policy for DeepSeek Harness: a host allowlist on web_fetch and web_search enforced at connect time, audit-mode by default, with OCSF Network Activity records
StateGraph orchestration engine as a DeepSeek Harness plugin: declarative directed state graphs with conditional routing, loopback, iteration guards, and a trajectory event stream.
Request image optimization, intelligent retention, and observability for DeepSeek Harness (DSH)
Performance diagnosis, repeated plugin isolation campaigns, safe recovery, and measured verification for DeepSeek Harness
DeepSeek Harness (dsh) Windows 环境防坑插件：两个防坑 skill（windows-enc/windows-sys：编码/转义/路径/进程/乱码预防规则）+ 三层主动防护（pwsh 结果乱码检测提示、危险写命令拦截、windows_encode_detect/fix 编码诊断修复工具）。零运行时依赖，零构建。
DeepSeek Harness plugin: DSH-native righthand tools — durable KV store, credential/settings management, governed exec, and a tool guard
DeepSeek Harness 工具调用安全守卫插件：规则引擎拦截 + 路径白名单 + 敏感文件黑名单 + 人工审批 + 审计日志 + 侧边面板
Cloudflare D1 tools for the DeepSeek Harness: d1_list/query/exec/schema/stats/health over the D1 HTTP API — read-only first, lexical read-only guard, write-approval gate, row clamping, CSV/JSON output.
Execution-fidelity guard bundle for DeepSeek Harness
Intercepts `rm` commands issued by any DSH agent session (GUI, automation runs, headless bridges) at the tools guard layer and moves the targets to the macOS Trash instead of deleting them. Switchable from the web GUI under Settings → General.
DSH 危险命令守卫：通过 tools/pre-execute 门禁拦截 bash/pwsh 高危命令（rm -rf /、mkfs、dd 写设备、管道执行脚本、git push --force 等），支持分级策略（极高危始终拒绝 / 中危跟随全局 deny/ask/log）、用户自定义规则与白名单。
Remote SSH terminal for the DeepSeek Harness web UI: main-area terminal tab, dangerous-command guard (open/guarded/locked), confirmation workflow, and model tools (netshell_servers / netshell_run). Passwords live in the DSH encrypted credential store and never pass through AI sessions.
DeepSeek Harness API 余额守卫：底部状态栏实时显示 API 余额（¥），低于自定义阈值时暂停模型输出并引导跳转充值页。
Local, zero-network, fail-closed secret preflight for DeepSeek Harness model requests.
Local, deterministic release risk checks for DeepSeek Harness.
Restart recovery for dsh web: after a crash/restart, an interrupted agent turn continues automatically (host-side agent/created listener — no browser timing races). Pairs with the dsh-web-guard skill (which auto-relaunches the web process); together they
dsh bundle for computer-control: desktop control (screen capture, pointer/keyboard injection, accessibility-tree semantic actions, safety guards) bridged to the harness over stdio JSON-RPC.