Multi-Core Collaborative Novel Writing — a DeepSeek Harness (DSH) preset plugin that installs a multi-agent novel-writing mode: one conductor plus five specialized subagents (多核协同写作模式).
Agent 협업
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:sailoumili/novel-writer
DeepSeek Harness plugin: bridge local dsh web to the OneBox Android app over the public internet (scan-to-pair, E2E encrypted tunnel). 万宝盒 App 的 DSH 云端中继桥插件。
Agent 협업 · 서비스 연동
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:wangzhishou/onebox-dsh-bridge
A structured multi-panelist review workflow for the DeepSeek Harness: N panelists score in isolated subagents blind to each other, a chair reconciles disagreement, an independent critic audits the process.
스킬 · Agent 협업 · 개발 도구 · 생산성 도구
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:LeslieWylie/review-workflow
SHOPLINE AI Toolkit for DeepSeek Harness (DSH): bridges the official SHOPLINE Developer MCP server and ships SHOPLINE agent skills, mirroring the Shopify AI Toolkit architecture. dsh-plugin.
스킬 · Agent 협업 · 개발 도구 · 서비스 연동
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:lunw/shopline-ai-toolkit-dsh
Hooks that keep a coding agent's green claims honest: verification runs are recorded unmasked, and "done" is blocked when the evidence is stale, failed or masked.
Intent fitting for DSH: the agent predicts several directions first, asks a batch of clarifying questions each round, narrows the distribution until one converges, then proposes it for your confirmation. Bottom-right floating panel, local JSONL archive, and both context injection and memory writes behind independent opt-in switches. 记忆拟合:AI 先预定若干方向(预测你的意图),每轮批量提问,轮间思考并展示方向可信度,收敛后提案待确认。右下角悬窗 + 本地留档;上下文注入与记忆写入各自独立开关,默认关闭。
메모리 및 지식 · Agent 협업
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:Aik358/dsh-memory-fitting