Blog
In-depth articles on AI agents, LLM engineering, servers, Python and software development.
dsh-answer-pet is a DSH Web answer status pet framework (MIT, 0.6.0) maintained by Nanki-nn. The pet switches animations between thinking, outputting and tool calling. Its status card summarizes token
DSH-Vision-Proxy is an MIT-licensed DeepSeek Harness interface enhancement plugin maintained by Flyvhidbwo. It registers the `deepseek-vision` route, converts GUI-attached images into text via an Open
The `dsh-undo-plugin` is a DSH community plugin maintained by lire1131, with the package name `dsh-undo-savepoint` after installation. It automatically and manually takes snapshots for configurations
# dsh-skill-manager is maintained by sulfide2085 under the MIT License. It adds a skill management section to the DSH Web settings page, merges skills from the DSH / Agents / Codex / Claude directorie
### Maintained by gxpppp, dsh-search-mcp replaces the built-in `web_search` backend of DeepSeek Harness with search-class MCPs, while keeping the tool name and display unchanged. This article verifies
dsh-record-replay is a DeepSeek Harness plugin maintained by humblebanana that integrates Open Record/Replay into the runtime: it uses the `orr_*` tools to record macOS desktop demonstrations, verify
dsh-ramify is a DeepSeek Harness community plugin maintained by yanglongyun, which uses a tree-shaped canvas to generate, compare and iterate multiple interactive solutions. This article checks agains
### Introduce the community plugin dsh-plugin-yet-another-subagent: Use a subagent tool with profile parameters in DeepSeek Harness to manage multiple sets of subagent strategies, and provide Web UI f
`dsh-plugin-development` is an MIT-licensed open-source Agent Skill maintained by w2112515. It is used to design, implement, package, review and diagnose DeepSeek Harness plugins, with an optional bun
dsh-playwright-browser is a community plugin for DeepSeek Harness maintained by Clizo1209. It provides 10 browser_* tools, semantic positioning and multi-tab control for agents via Playwright. This ar