The web search plugin for DeepSeek Harness, and the search bridge for every model without native web access. Ask the web or X, get structured JSON evidence. | 🥇 全网最强的 DeepSeek Harness 联网搜索插件🥇。为不能联网的模型补上搜索。问网页或 X,拿回结构化 JSON 证据(搜索、抓取、引用)。
Model-driven context management (Active Context Pruning / ACP) for the DeepSeek Harness — the model decides when and what to compress. Ported from billion-context-pi (ranxianglei); acp-kernel reused verbatim. CompactionEngine backend with compress/decompress/search_context/acp_status tools.
Interactive long-session background agents for DeepSeek Harness: start a durable continuable child agent, watch its progress in the Web UI sidebar, message it any time, and interrupt it - all through the official subagent seam.
Local personality, memory and distillation layer for DeepSeek Harness — SOUL/IDENTITY/USER/STATE persona files, cross-session memory and LLM distillation, auto-loaded on startup.
Retrieved ≠ injected: CBDC-gated memory for DeepSeek Harness — decides how memory is USED (use/verify/ignore + feedback learning + audit), not just stored
Unified fleet-wide agent memory system for DeepSeek Harness — shared Obsidian Vault for dsh, Codex, Claude Code & Hermes with dependency-free Python FTS5 core.
Multi-agent collaboration suite for DeepSeek Harness: specialist roster with on-demand dispatch, roundtable, model comparison and a multimodal vision bridge — models via the official provider flow.
A formal workflow protocol that keeps AI coding agents on track through machine-enforced invariants, verified evidence, and structured context. Designed to keep deep, long-running projects coherent across sessions, agents, and months.