Plugins

Splits New Session in the DSH web sidebar into two modes: a 日常聊天 row opens a session on the `daily` agent preset the plugin installs into the harness home (persona, web search and fetch, ask-user and compaction — no files, shell, plan mode or subagents), inside its own daily-chat workspace whose sidebar region lists only those chats, while 开始工作 keeps the shipped coding agent; the built-in New Session button is hidden and every other New Session entry point follows the active mode, which is remembered in localStorage.

GLM Mode agent preset for zhipuai/glm-5.3-flash: PTC/Code Mode tool presentation, a reasoning loop guard and a tool spin guard (live-incident driven), and GLM-specific auto-compaction at 75%/22%.

人脑式上下文压缩逻辑，大幅减少长任务上下文占用以及token消耗

Auto-resume for interrupted agent sessions: an ordered rule table routes by failure type (rate limit, quota, auth, context overflow, crashed orphan) to automatic backoff retry, model switch, resume after context compaction, or a stop-loss notification, with a visual rule editor and a full activity log.

A Taffy mood mascot for the DeepSeek Harness Web UI: the corner GIF reacts live to agent state — approval pending / approved / rejected / begging, thinking, tool runs, interrupts, ask_user_question waiting, compaction, long-task fatigue, idle-ignored approval, idle / sleeping — with SSE-pushed state, drag positioning, middle-click console, tool-card mirror console, per-mood locking, persisted appearance settings and per-day bucket statistics.

Analyzes conversation images through mode-specific prompts (caption, UI, document, grounding, topology, etc.) and injects structured evidence with coordinate primitives (boxes, points, refs) as text, with session-level caching for reuse across replay and compaction.

Vision toolkit for text-only DeepSeek: model-invokable `vision` tool, wrapper adapters for deepseek/opencode-go (v4 flash/pro), Antigravity IDE quota (default, flash/pro) / any OpenAI-compatible VLM / Gemini / local Ollama channels, evidence memory with compaction rehydration, content-hash cache, and a bilingual client panel.

DSH plugin: counts completed compactions per session and raises a floating reminder — with bilingual UI — only after repeated compactions plus heavy extra token spend.

Windowed, reversible context management for DeepSeek Harness; replaces Basic compaction.

Compresses older DSH context into a structured handoff, retains recent messages, and keeps session history searchable.

Compaction fallback for compaction-basic: when its LLM summary throws, comes back empty or cannot fit the window, the span is compiled deterministically with a seq pointer on every line (vendored dsh-compaction-instant compiler). Adds recall and search tools, /recall and /context commands, an idle trigger at 45% of the window and an on-disk span archive. The engine mounts as one row in the agent preset (docs/dsh-integration.md).

Model-autonomous context compaction for DeepSeek Harness: a mind_map tool surveys the agent context surface (stable seq ids, roles, token weights, clear-boundary markers, the clear-mind playbook), a clear_mind tool replaces a chosen conversation-history span with a model-written checkpoint through the native compaction transaction, call and result pairs self-collapse into one-line tombstones, and a proactive reminder nudges the model when the context grows long or a turn runs too many steps.

Injects the rule files under $DSH_HOME/rules and <cwd>/.dsh/rules into the conversation at the first agent pre-step, re-injecting them after compaction.

Context curation for DeepSeek Harness: content-aware tool-output slimming, dead-log sweep, pinned user requirements and in-flight work, archive-then-compact with an audit trail.

Per-provider concurrency gate on the llm/stream waterfall: holds surplus model requests in a FIFO queue inside DSH so main agent, subagents and compaction take turns on a single-slot local server (llama.cpp --parallel 1) instead of timing out; maxConcurrent / maxQueued / queueTimeoutMs.

Compaction quality / hallucination detector: when a conversation span is compacted, check on two axes that key information survived 鈥?FIDELITY (assertions in the summary must be found verbatim in the original leaf events, else flagged as fabricated) and C

Step-0 observation probe for dsh-compaction-audit: logs every compaction/* session event and probes whether shadowedSeqs still resolve to readable events in session.events. Observe-only; never injects, never blocks.

DSH web search and Responses native V2 compaction for Sub2API-proxied or NewAPI-relayed GPT models

Adaptive Reversible Context (ARC) for DeepSeek Harness — provider-aware, cache-conscious context governance with reversible recovery.

One-command bundle of the three AgentFrame plugins: memory-director, compaction, and aura-scheduler together.

Routes chat images to a fixed OpenAI-compatible vision model, returns factual observations to the selected main model, and reuses session-scoped observations across replay, compaction, and restarts.

Execution flow chart: renders the current session's execution as a vertical node flow (turns, user, assistant, tools, approvals, retries, compaction), appended in real time over SSE with auto-follow scrolling — the chart draws wherever the session executes.

Replaces the built-in DSH compaction summarizer with a five-section handoff summary and keeps every shadowed message readable in a segment directory.

Tiered compaction hub: replaces the single 80% compaction cliff with a configurable ladder (default 30/50/70/90) whose tiers escalate from gentle to maximum consolidation laws, folds a deterministic keyword index into every checkpoint so facts accumulate across cycles, and summarizes through a model chain — primary plus up to 3 fallbacks, each with its own reasoning effort. Adds a Compaction module sub-tab and a Settings card; toggling off always restores stock compaction.