context-engineering

Cross-agent plugin for manually excluding and editing AI conversation context, with search, filtering, hide/restore, undo, and original Session preservation.

让 DeepSeek Harness 拥有跨会话长期记忆：AI 自动记住你是谁、你的项目和偏好，新会话直接带上背景，生活与工作记忆自动分开互不干扰，零配置无感运行 | Long-term memory for DeepSeek Harness: the AI remembers who you are, your projects and preferences across sessions, keeps personal and work memories separate, zero-config.

DeepSeek Harness 的 AdaptiveContextPlane (ACP) 插件——带治理的长期记忆系统

Read-only performance diagnostics for DeepSeek Harness: session load/restore timing, spill-hit counts, compaction count and trigger, context-injection volume (AGENTS.md/skills/tool-schema token share), and LLM cache hit rate — surfaced via /fast, persisted as reconstructable session events with async sampling off the model path.

Cut large agent tool output by what it means, not by where it was cut. 13 content-aware reducers + DeepSeek Harness plugin.

DeepSeek Harness MCP tool search for large catalogs: 1,000 MCP tools behind 2 MCP-facing schemas, exact-schema calls, allow/deny controls, and a local calculator.

Execution-time drift firewall for long-running DeepSeek Harness agents. Real-Harness tests: unsafe stale mutations 12/12 native -> 0/12; valid controls 7/7 both; post-SIGKILL unsafe continuation 2/2 -> 0/2.

Pass your project, not your context.

X-ray for your DeepSeek Harness: context-tax attribution, entry inspection, dependency cascades

Shrink massive MCP catalogs to two tools, letting DeepSeek Harness search and call 1,000+ remote APIs efficiently.

The navigation of AI Agent.

Memory and token-optimization plugin toolkit for DeepSeek Harness: cross-session knowledge graph memory + five-layer token-saving orchestration.

Memory plugin with built-in curation and governance for DeepSeek Harness: 为你的 DSH 赋予可塑性记忆

Evidence-preserving tool output compression for DeepSeek Harness — deterministic, local-first, and benchmarked.

Evidence-backed working memory for DeepSeek Harness: a cited ledger of what already worked in this workspace, built from the session log you already have. No index, no model, no capture step.

Markdown folder long-term memory plugin for DeepSeek Harness (dsh): one file per fact, human-editable, git-versionable

Local-first continuity plugin for Codex and DeepSeek Harness across compaction, resume, and handoff.