Plugin

Vision provider route that transcribes attached images to text through a configurable model (15+ OpenAI-compatible and Anthropic vendors) while DeepSeek keeps answering.

Two-tier image reading for text-only models: fast local OCR (RapidOCR, offline) first, vision-model fallback (modlens).

Give DSH text-only models vision: an image/OCR/document recognition skill (race pool → custom channels → local) plus an idempotent host patch so image messages reach the model.

Ingestion-time tool-output curation: keeps error and warning lines plus head and tail, spills the full text to disk with a pointer the model can read back; covers every text tool, small results pass through untouched.

MCP server manager: side-panel with per-status server listing, stdio/streamable-http setup forms, mcpServers JSON import, exponential-backoff reconnect, and tools registered to the model as mcp__<server>__<tool>.

Notes and todos as plain Markdown files in your workspace: type /ml plus a sentence to jot into today's journal; todos come in three flavors — deadline, sleep-until-a-date (auto-wakes when due), and anytime; /ml view fuzzy-opens any file with a few keystrokes. Zero tokens throughout — nothing goes through the model, and your notes stay ordinary local files that git and any editor can read.

Observational file-mutation rollback for the fs tools: write/edit pre-images are checkpointed as git blobs or snapshots and restored through a model-facing rollback_files tool or a /rollback command.

Chemical safety intelligence through the hosted MSDS Chain MCP endpoint: 23 tools for compatibility and mixing order, GHS hazards, PPE, storage, waste disposal, exposure limits, transport classification, multi-region regulatory compliance, SDS lookup and version diffs, and signed audit reports, each answer citing the supplier SDS and revision date it is grounded in.

Manages multiple Jenkins servers and triggers jobs from a Settings page, model tools, or a per-workspace "Run Jenkins Job" entry, with a bilingual host + browser UI.

Auto-discovers model listings, pricing and capabilities from OpenAI-compatible API hosts, normalizes them, and emits ready-to-use model configs for dsh.

Persistent model tool-call trace recorder: every tool call is durably written to a JSONL file that survives restarts, queryable via a structured call_trace tool and a callTraceHistory service, with size rotation and an optional floating canvas UI add-on.

OpenAI-compatible image generation for DSH Desktop: the model calls generate_image in chat and the images render inline in the conversation.

Zero-dependency MCP client bridge that connects stdio and HTTP MCP servers and registers their tools as mcp_<server>_<tool> for the agent.

Auto-supplement QMD semantic search when the model greps/globs a knowledge-base directory: zero-dependency DSH plugin, async injection via next-step, no manual invocation.

Tracks the agent's state as Yogācāra mental factors over a durable seed store, and feeds it back into the prompt as first-person inclination.

Bi-temporal memory backed by a SodaMem daemon: recall is contributed to the prompt while each turn is assembled and every closed turn is ingested, with no tool call and no model call on the read path; each recalled fact cites the turn it came from and carries validity dates, so superseded facts stop being retrieved.

LLM-driven remember/forget memory: after each turn a model decides what to keep (MemoryDirector), injects relevant memories before each step, dedups, and persists cross-session.

Project-scoped persistent key-value memories with set/get/list/delete/search model tools and JSON-file storage.

Focus board for DeepSeek Harness agents: durable, model-maintained notes in the session workspace that pin the objective, constraints, and decisions across compaction and sessions, with automatic context injection, archive on clear, and an optional web panel.

Adds a configurable humanized whale-girl maid persona with immersive roleplay, speaking modes, and a lightweight layered memory with query-triggered recall.

Two-phase session captioning: instant keyword titles while the conversation is busy, then cheapest-budget-model refinement plus a one-line summary once it goes idle.

Automatically fills in reasoning-effort levels for models from unofficial (custom) providers, with data sourced from models.dev.

A GUI subagent-model selector beside the main model seat: pick a model and reasoning effort per session, applied to every in-process subagent.

xAI Grok subscription chat: SuperGrok or X Premium+ OAuth, a displayed model subset, Responses through the Grok CLI proxy, and always-on server-side web_search and x_search.