Plugins

Adds a download_file tool that fetches a URL over 16 aria2c connections and falls back to a single stream when aria2 is unavailable. Needs aria2c installed separately; the aria2 channel does not verify TLS certificates.

Local memory for DeepSeek Harness: what you write is filed under a topic and linked to related entries, recall ranks results by embedding similarity (word matching when no model is installed), and the data stays in a local SQLite database.

Bridges DeepSeek Harness to a Hindsight memory service using the same conventions as omp: one shared bank, automatic project-tag stamping, project plus global recall, plus tools for mental models, duplicate detection and retrying failed background operations. Requires a running Hindsight server (default http://127.0.0.1:8888).

Turns a local Markdown vault into two folders the user picks in Settings: a knowledge folder DSH writes notes into, and a memory folder used as the agent's long-term memory with per-turn recall and end-of-turn capture. Pure Markdown with frontmatter, BM25 search, reversible archive, and path isolation that rejects traversal, absolute paths and symlink escapes.

Runs the upstream Headroom compression toolkit (headroom-ai, Apache-2.0) inside DeepSeek Harness over MCP: headroom_compress, headroom_retrieve and headroom_stats are exposed as mcp__headroom__* tools backed by the upstream server, installed on demand by uvx. Compression is lossy but reversible — originals are fetched back by hash from a local CCR store. Also ships a registrar that adds DeepSeek Harness to Headroom's own MCP registry, and a model-header sanitizer for the upstream proxy.

Semantic context watcher for DSH agents: maintains a journey ledger with flash summarization and MinHash novelty detection, folds milestones, and can inject long-term context with a three-level stall response; injection and the stall-2/3 responses are off by default.

Local-first knowledge base for DeepSeek Harness: imports Word/PPT/Excel/PDF (text or scanned), EPUB and MOBI/AZW3, builds Markdown parts with page anchors, OCR quality/confidence and column-aware PDF reading order; adds an LLM-maintained wiki layer (schema.md rules, auto index.md, append-only log.md, typed notes with sources and backlinks), adds source-trust labels (high/medium/low) with trust-aware ranking and low-trust warnings, and supports ranked search, cross-KB query, precise reading/editing, and md/jsonl/zip export. MOBI/AZW3 need Calibre; legacy .doc/.ppt need Word/PowerPoint (Windows) or LibreOffice (Linux); scanned-PDF OCR downloads models on first use.

Lightweight crash-recovery for dsh: after a restart, it detects sessions whose last turn was killed mid-flight and re-injects a continuation prompt that first demands the agent verify real-world state before acting, so side-effecting operations are not blindly repeated. It only reacts to the genuine crash marker (turn/end reason 'interrupted', synthesized on cold read by session repair), so it never misfires on user cancels, rate limits or model errors - which is why it needs no error classification, backoff or configuration UI. Unlike general-purpose auto-continue plugins, it also covers the case where a subagent was interrupted mid-run: the parent turn has already ended normally while it waits, so a naive check sees nothing to resume and the whole subagent tree dies silently; this plugin detects the interrupted child and wakes the parent instead. Single host file, zero external dependencies, seven config constants.

Select a statement in an AI answer, find candidate support in earlier chat and tool records, classify support, and jump to original passages for review.

Purges billion-context snapshots left behind after a session is deleted in DeepSeek Harness, and can also clean up cost-meter ledger rows (off by default).

Bridge DSH to OpenCode-compatible models via the local OpenCode CLI: native LlmAdapter passthrough, persistent OC sessions with cross-turn tool results, tool mirroring, and a CLI environment check panel. Requires the OpenCode CLI installed locally.

LocalAI provider for language models, embeddings, reranking, speech, and image generation with LocalAI-specific controls in DSH. Requires Firefox2100/dsh-ai-providers#runtime to be installed and enabled.

Jina provider for embeddings and reranking in DSH. Requires Firefox2100/dsh-ai-providers#runtime to be installed and enabled.

Cohere provider for language models, embeddings, and reranking through Cohere's native APIs in DSH. Requires Firefox2100/dsh-ai-providers#runtime to be installed and enabled.

Anthropic provider using the native Messages API for language models in DSH. Requires Firefox2100/dsh-ai-providers#runtime to be installed and enabled.

Opening animation, three bundled wallpapers, up to nine uploaded wallpapers with random rotation, and live panel and sidebar opacity.

Dark-only starry-sky theme for the Web UI: both Light and Dark resolve to one deep-space night palette with a twinkling click-through starfield and four constellation figures, and the appearance preference is pinned to Dark while the plugin is enabled.

Distill over-long model reasoning into its useful core, replay the compressed version on every later request, and report each session's cost and savings. Disabled by default: select one rewrite backend first. derive needs no install and works on packaged DSH; in-place needs 14 source patches and a rebuild (recommended for source checkouts); message-edit needs a third-party plugin and is experimental only.

Token usage dashboard and per-session tree statistics: provides time-range filtering, 6 summary metric cards, GitHub-style activity heatmap, model stacked-bar charts with cache hit rate curves, and Top-5 model/provider rankings; enhances the native session status bar to display aggregated token usage across main agents and all subagents with detailed breakdowns.

Windows only, needs the NetEase Cloud Music client installed. NetEase Cloud Music player for the DSH Web UI: a mini bar that expands into a liquid-glass card with search, synced lyrics and volume, a click-to-open spinning vinyl view, and it signs in with the local NetEase client account (cookie decrypted on this machine only) so VIP tracks play and the client current song can be pulled in.

Sparkle button immediately left of the model selector that rewrites the composer draft into an executable prompt in place, modelled on WorkBuddy: the host calls the conversation LLM with a prompt-engineering system prompt, multi-level undo keeps up to 8 previous drafts, and a token-level diff card shows added and removed spans, the number of changed spots and the net character delta. Runs on the session model, is disabled while the draft is empty, and keeps the browser-session cookie fence in front of the LLM call.

Turns one plain-language request into a live workbench window: a sidebar panel asks a few clarifying questions, the model writes a single-file HTML artifact, and the plugin mounts it in a resizable floating window or a standalone browser tab. Declared window actions are recorded as info-level project journal entries (without their payloads), and workbenches can be imported, exported, or installed from a market index.

Adds an installed-plugin picker beside the composer: drag a plugin name into the prompt at the drop point, open a draggable plugin rail in the right sidebar by double-clicking the entry, and choose which plugins (and remarks) appear in Settings.

Interactive cards in replies, built from a fixed catalog of 31 components (sliders, inputs, sortable tables, charts, function plots, typeset formulas, tabs, checklists, flowcharts, quiz buttons, an animation player): controls recompute in the browser without another model call, and the model never writes HTML or JavaScript. Includes a calculus engine (integrals, derivatives, limits, sums, roots, Taylor polynomials) and a bundled skill that teaches the model when to use a card.