Plugins

Curated file-based long-term memory for DeepSeek Harness — human-readable markdown notes, one plain-file store shared across hosts (git-ignored when it lives inside a project repo), and a deterministic index cap enforced by ctx.tools.guard()

Skills and optional MCP rows for eleven marketplaces: price comparison across Wildberries, Detsky Mir and Yandex Market, plus per-source search, product cards and reviews, including Ozon, Avito and AliExpress. The 14 skills load on install; both MCP rows stay disabled until RU_MARKETPLACE_MCP_DIR points at a local clone, which needs Python 3.12+ and uv.

Bounded, layered, approval-gated, auditable cross-session memory: a typed `ctx.memory` seam with a zero-dependency SQLite provider, a `memory` tool, and frozen snapshot injection, plus the dsh-memory-protocol v1 rehearsal with an adapter registry and a distributable conformance suite.

baro as a DeepSeek Harness subagent provider: delegate a multi-story goal, get parallel execution, independent review and a verified outcome back — with a live run panel in the sidebar

Code-first React and TypeScript task apps rendered inline, in Canvas, full screen, or on localhost, with interaction state available to later agent turns and approval-gated MCP and API access.

Superpowers for DeepSeek Harness: TDD, debugging, planning, and collaboration skills adapted from obra/superpowers, installed as a Cordis plugin registering 14 skills with zero runtime dependencies.

Claude Code-style auto-memory, natively: typed memory files (user/feedback/project/reference) with a MEMORY.md index auto-injected into the system prompt (zero footprint when empty); memory_write/read/list/delete follow Claude Code write rules (dedupe-and-update, absolute dates, cross-links); workspace-scoped and user-scoped layers, cross-process file locking with orphan-lock recovery, plain Markdown on disk, no external services, no embeddings.

Adds account login, API and custom Provider setup, model switching, image fallback for text-only models, and token/cost attribution to DeepSeek Harness 47f.

Bloom Morandi theme: ten variants (mist, cinnabar, petal, ripple, sage, stone, lapis, amber, aurora, lavender) built on OKLCH so light and dark stay perceptually even, switchable from the top bar, with all 20 foreground/background pairs measured at WCAG AA or better.

Red-team engagement mode: send one target organization name and a planner session runs a five-role execution team (recon, asset triage, vulnerability discovery, exploitation, internal pivot) under a user-configurable concurrency limit (three by default, changed from the console and taking effect immediately), preflight-checking skills and resources and asking once for any missing key or VPS first, guided by a built-in first-run onboarding skill; scoring follows the merged intrusion scoring rules (8 categories / 25 points) with server-side caps, highest-privilege-wins and per-service dedup, and self-registered accounts never score — a login that fits no more specific category is scored as controlling a web application system rather than dropped; findings land in a local SQLite fact base with discovery timestamps, shown in a persistent right-side console of 9 tabs that follows the selected session — switch sessions and it shows that session's target organization, resolving sub-agent sessions to their root session and falling back to the session title, the console footer always naming which of the two matched (its per-session binding file, or the title); (asset mapping with a discovery timeline and the live test view, the agent roster with its concurrency setting and editable role prompts, session/tunnel state, findings — vulnerabilities, credentials and access sessions plus the attack files archived against each vulnerability — five-stage attack chain, scoring targets, an evidence-screenshot capability (register a screenshot per score point, at most three representative shots per point so a bulk result such as dozens of endpoints does not flood the console or the report, viewed inline in the console and embedded in the report, with on-site screenshots kept strictly apart from record-generated evidence cards), and a report that spells out how each score was obtained (actions, exact commands, credential provenance, tunnel build commands, a reproduction entry that is either a paste-ready HTTP request or a runnable command, plus success criteria and an evidence index), a category-organised POC/EXP knowledge base, and a skill library that reports for every unusable skill why it is unusable and the concrete steps to repair it); ships 14 reconnaissance/assessment skills (the npm package deliberately carries no tool auto-installer and no attack-chain skills — those ship as a separate toolkit Release attachment that the operator fetches and runs themselves), 56 redteam_* tools and one-command self-update.

Causal memory for dsh: a zero-dependency MCP stdio bridge that mounts the 17 tools of the causal-memory server under clean names (no mcp__ prefix) — decision → outcome edges with typed relations (caused / enabled / prevented) in a local SQLite store that survives compaction — plus a pinned system-prompt section. Requires the causal-memory server binary on PATH (pip install causal-memory, or a prebuilt release).

Research-assistant plugin suite (literature search, research wiki, LaTeX compile, independent subagent review) for the DeepSeek Harness

Prevents request context bloat in DeepSeek Harness: catalogs MCP tools and skills under ctx.capability, exposes them in exposed/progressive/blocked tiers, and invokes low-frequency ones on demand via meta_search/meta_invoke, with a visual settings tab.

Execution maze for DSH sessions: the agent's real path — main advance, failed or dead-end detours, backtracks, request retries and subagent branches — drawn on one wall-clock timeline, with three tracks under each lane sharing that axis (tool-call density, token pulse, context pressure) and a deterministic analysis panel below it (tool result matrix, failure-recovery chains, duration distribution); watch it grow live in a session tab, or upload up to 5 session logs for a same-axis compare.

Drop-in enhanced fork of the official todo_write tool: priority tags on each task, full checklist rendering in the conversation, and a draggable in-session floating task panel with position memory, double-click reset and a live indicator of the action currently being executed.

Fluid streaming rendering and smooth scrolling for the DeepSeek Harness Web UI.

Renders agent HTML output as real UI in the DeepSeek Harness web GUI, with visual cards, KaTeX math, Mermaid diagrams, built-in open-source fonts, and design-spec injection.

Persistent enhanced web search: multi-engine routing (DeepSeek/Exa/DDG/Bing/Jina + GitHub/Bilibili/YouTube/V2EX/Xiaohongshu/Twitter/Reddit/RSS), SQLite+LRU cache, userscript-style extraction, Playwright rendering.

Word alternatives and underscore-gated fill-ins in the composer. End a line with _ and it is filled; misspellings are flagged as you type. Routes through the model dsh is already configured with, so it needs no API key.

Floating DSH desktop pet with 12 authored state animations for requests, tools, compaction, archive and prompt enhancement; live context occupancy; a persistent cross-session Lifetime Ledger; per-provider/model totals, hourly trends and 7/30-day views; dark/light theme switching; optional cost estimates (off by default) with a visual price editor and monthly budget alert; client-side skin ZIP import; JSON/CSV usage export; snapshot-only panel reads with background incremental indexing; and opt-in prompt enhancement with editable preview and DSH composer submission.

Personal assistant for DeepSeek Harness that turns conversational commitments into cross-session plans, reminders, and an auditable dashboard.

DSH notification & remote-control plane: one `notify()` API across 28 outbound channels, six inbound control channels, phone approvals/questions/task control, a Native DSH Notify & Control UI, live outbound hot-apply, multi-agent routing, and zero runtime dependencies.

Puts the Markdown documents, source files, images and video that already exist anywhere in the session workspace into the DSH sidebar, ranked by relevance to the current conversation. The default lexical mode matches your recent messages against title, summary and body locally with IDF weighting (BM25) — no model calls, no network, nothing leaves the machine. The list comes from scanning the workspace rather than from a recent-files list, so restarting DSH or opening a new session does not empty it. Results are organised into a three-tier reading order (primary / supporting / related) with a verifiable reason on every row — which field matched, which document cites this one — and one click opens an in-place preview: Markdown as rendered text, relative-path images actually drawn, source code with the host's own syntax highlighting, video streamed over HTTP Range. The same Context Pack is exposed to the agent as a knit_docs tool. Later versions added read evidence and lifecycle tracking (attribution is frozen at the moment of reading), pin / exclude (exclusion applies before retrieval, and is undoable), relation evidence (references / imports / tests, with line numbers) and an incremental index reused across restarts. Since v0.23 there is also an optional hybrid mode: it is off by default and only runs once you configure an Ollama or OpenAI-compatible embedding provider yourself. When it runs, what leaves the machine is your query text plus each candidate file's title and summary — a Markdown summary is an excerpt of the body's first paragraph (at most 60 characters), a code file's summary is its workspace-relative directory, and an image or video contributes its file name; full bodies and absolute paths are not sent, and a loopback provider keeps everything on this machine. Since v0.24 it also associates uncommitted changes: a read-only git status (fixed argv, no shell, whitelisted child-process environment) combined with the same reference / import / test / mention evidence shows which files a change is likely to drag along, and it is off by default. The benefit of semantic retrieval still awaits a real-provider run, so no recall improvement is claimed.

Host page for kit plugin configuration in the DSH Web GUI: registers a "Plugin configuration" section in Settings — peer to the built-in General section — through the official settings.section slot and declares a settings.kit.item child slot, which dsh-plugin-kit plugins render their own config cards into. Installing it alone shows an empty page; cards appear as slot-registering kit plugins are installed.