context-engineering 相关的 DSH 插件

thoughtdag

An infinite canvas where LLM conversations grow into an editable thought graph. Read papers into maps…

leantoken

Code intelligence for agents: find the code that matters and keep your context window and tokens lean.

dsh-auto-memory

Auditable memory for DeepSeek Harness: 4-state trust machine + verbatim evidence chains + recall funnel + 4-layer evals.…

dsh-sieve

Token-saving judgment plugins for DeepSeek Harness

dsh-engram

Local-first cross-session long-term memory for DeepSeek Harness: dual-scope SQLite stores, hybrid FTS5 + local-vector re…

dsh-multitask

Multitask 减震器模式（Agent preset）：主对话只做需求理解、任务下发与结果汇总，读文件/跑命令/构建/搜索等污染性工作全部下放子代理…

agent-body

Agent-Body — an organ-based plugin layer for DeepSeek Harness: organs, nerve impulses, a heartbeat, reflex arcs, long-te…

dsh-harness2

Harness² (recursive agent harnessing) for DeepSeek Harness: Propose–Probe–Compose self-improvement as a DSH plugin.

dsh-domain-role

Domain-identity framework for long-running AI agents: one domain, one long-lived identity, one accountable owner, one au…

session-watcher

Local Claude Code context-cost monitor, transcript replay, buckets, and handoff