NanoBot Subagent Bridge — lets DeepSeek Harness run the local nanobot AI agent (HKUDS/nanobot) as a subagent via a nanobot_run tool (oneshot CLI or persistent OpenAI-compatible server).
Agent 협업 · 서비스 연동
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:2025Bigeye/dsh-nanobot-subagent-link
Notion block conversion for DeepSeek Harness: turn an exported block array into Notion-flavoured Markdown and back, with rich-text runs split at annotation boundaries and every lossy step reported. No API client and no external dependency.
서비스 연동
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:yuehancn/dsh-tool-notion
Podcast tooling for DeepSeek Harness: detect chapter boundaries from silence, cut a recording into segments, write an iTunes-compatible RSS feed, and normalise loudness. Shells out to a local ffmpeg.
서비스 연동
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:yuehancn/dsh-tool-podcast
Encode and decode QR codes from DeepSeek Harness, with symbol-capacity planning: pick the error-correction level, check the payload fits, and know the printed size needed for a scanning distance.
서비스 연동
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:yuehancn/dsh-tool-qrcode
DeepSeek Harness plugin: per-tool call-count and result-byte caps, enforced before the body runs and after it returns, failing loud rather than truncating
Agent 협업
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:jwilson411/dsh-tool-quota
Tool-result dedup policy for the DeepSeek Harness — replaces byte-identical repeated plain-text tool results with a pointer to the first occurrence (hash-indexed), saving input tokens on long sessions.
Zero-loss tool-result pruning for the DeepSeek Harness (DSH): oversized tool results are spilled to a file first, then pruned with exact elided-span offsets and a recovery locator. A dsh-plugin.
기타 확장
$ npx -p @deepseek-ai/dsh dsh plugin --profile web add github:may3rr/dsh-tool-result-guard