@dsh-scholar/research-plugin
DSH Research OS — a fully automated scientific research plugin for DSH (DeepSeek Harness): survey, idea, experiment contract, durable runner jobs, claim-evidence ledger, manuscript and release bundle.
59 results
DSH Research OS — a fully automated scientific research plugin for DSH (DeepSeek Harness): survey, idea, experiment contract, durable runner jobs, claim-evidence ledger, manuscript and release bundle.
Connect DeepSeek Harness (dsh) to a local Obsidian vault: search, read, write, move, and trash notes.
Tencent Cloud Service CLS observability plugin for DeepSeek Harness — direct upload to Tencent Cloud CLS
Langfuse observability for DeepSeek Harness: OpenTelemetry traces, compaction, feedback Scores, and fork lineage
统一 LLM 请求/响应检查器:调 reasoning effort、外部思考(think)导出、流量与包分析 —— DeepSeek Harness 插件
DSH 通用学习 Agent「学习伙伴」:文档式笔记插件——Obsidian vault 多根检索、文件夹规划与硬门禁落盘、微目录与覆盖度、学习进度与跨会话记忆
Three learning skills for DeepSeek Harness: course notes, conceptual explanations, and self-contained lecture chapters.
Obsidian 第二大脑(移植自 obsidian-second-brain)
DSH插件,以Observational Memory方式管理上下文
SLI/SLO/错误预算:定义可衡量的可靠性目标、错误预算告警、SRE 实践。受 wshobson/agents(38k★ MIT)启发。
事故手册:严重度分级/检测→分类→缓解→恢复→沟通流程/升级路径/值班交接。受 wshobson/agents(38k★ MIT)启发。
量化回测与风险指标:规避前视/幸存者偏差、考虑交易成本、walk-forward 分析、风险指标计算。受 wshobson/agents(38k★ MIT)启发。
事故复盘写作:无指责文化、根因分析、时间线、行动项,驱动组织学习。受 wshobson/agents(38k★ MIT)启发。
多 agent 团队协作:并行调试、并行功能开发、任务协调、沟通协议、团队构成。受 wshobson/agents(38k★ MIT)启发。
架构模式:Clean Architecture、Hexagonal、DDD 战术模式、依赖规则。受 wshobson/agents(38k★ MIT)启发。
微服务架构模式:服务边界、通信、数据管理、韧性模式、事件驱动。受 wshobson/agents(38k★ MIT)启发。
Installable RouteCertificate advisory observer bundle for DeepSeek Harness users.
Content-addressed readiness proof for sanitized DeepSeek Harness observations on managed Windows hosts
Vault 观测插件:Obsidian vault 元数据快照 + 编辑统计 + 观测面板(可配置任意 vault)
数据叙事:用可视化/上下文/说服结构把数据变成推动决策的故事。受 wshobson/agents(38k★ MIT)启发。
dsh-codex-observability
Monorepo 管理:仓库结构、依赖管理、构建缓存、CI 策略。受 wshobson/agents(38k★ MIT)启发。
PCI 合规:支付卡数据保护、安全控制清单、合规评估要点。受 wshobson/agents(38k★ MIT)启发。
Failure journal: every tool call that exits abnormally is written to a durable JSONL journal on disk, clustered by error signature, and readable back by the agent through a `failure_journal` tool.