Plugins

Two-way bridge between GNU Emacs and a running DeepSeek Harness session over loopback HTTP + SSE: submit prompts, stream replies, manage sessions and workspaces, and answer ask-user/approval prompts from Emacs, with a Plugins-page card that installs the Emacs package.

Read-only Git panel for the current workspace — branch, changed files, diffs and image before/after compare — plus three capsules above the composer (Git status, tool and skill activity, session running time) and per-session change attribution with review progress.

In-session cd for DeepSeek Harness: re-base relative paths on the fly (the file tools and the bash/pwsh default workdir follow) or move the session to another workspace for real — a `cd {"workspace": true, "confirmed": true}` call the user has agreed to, or the user's own `/cd <directory>` command, relocates the session's stored log, sandbox root, AGENTS.md and skill discovery and sidebar grouping. Per-session, persisted across restarts, without moving the session for a plain re-base.

Keeps project decisions in plain markdown instead of chat history: one main document plus per-module documents, every entry carrying a file:line citation. The agent interviews you to turn open questions into settled decisions, and a five-dimension score (task complexity, extensibility, maintainability, code quality, reusability) is derived from the documents and shown on a button beside the model picker. A session can bind several projects; projects are sized small/medium/large, which widens the entry limits, and a reusable-pipeline section in the main document is injected when its trigger words appear.

Set an earlier automatic-compaction threshold per model on DeepSeek Harness: /autocompact writes a session-only override keyed by exact provider/model (off falls back to the backend defaults, default clears it), a Settings → Autocompact page manages the persisted per-model thresholds every session inherits, and a composer status entry shows the current reading.

Hierarchical workspace tree plus browser-style session tabs for the DSH sidebar: a "/" in a workspace or session title creates a virtual folder, with sections, pinning, time buckets, drag ordering, unified 6px status dots, per-row appearance customization (colour, glow, weight, icon) and a floating full-text search over session titles and message bodies. Several settings switches restore any behaviour to the host default, and a host-side route deletes sessions permanently (DSH ships archive only). Fork of KannaKuron/dsh-better-workspace.

Injects a small continuity pointer at the start of a new session, naming the previous session and any handoff file, and reuses DSH's own compaction summary when there is no handoff file.

Predicts the user's next prompt after each assistant reply and shows it as gray ghost text inside the composer; Tab accepts it, ↓ generates the next suggestion, ↑ returns to the previous one, and Esc dismisses it. Each assistant reply triggers an extra call to the session's model, carrying the most recent user and assistant text.

Read-only goal-tree status strip above the composer: planners see the full node tree, workers see their claimed node, and worker reports call back to the planner session.

Use Windows and native Linux DSH sessions in one Windows window, with WSL environment switching and bidirectional system tools.

Select text in the DSH web UI to stream a translation and an explanation of what it means in that specific context, read against the conversation it came from; the panel keeps a replayable history and can be promoted into a full session.

Behavior-drift reports for DSH sessions: compares two session logs and reports tool-sequence alignment, tool mix, reasoning volume, timing, retries and file targets, with an optional CI threshold gate. Zero dependencies.

Visual goal lifecycle for the DSH web UI: a composer dock to create, multiline-edit, pause, resume, complete and clear session goals with live round progress, plus a phase-dot composer entry and a settings-page goal overview.

Persistent Endeavour and Challenger peer sessions for DeepSeek Harness that plan, execute sequential tasks, and verify reported results.

Session-header task capsule with color-coded run states, click-to-jump and a multi-task dropdown, plus task notification cards (done / error / waiting-for-confirmation) with synthesized sound and optional system notifications.

Cron job board: schedules local scripts or dsh session tasks (fresh or pinned sessions) by cron expression, with inline timezone suffixes (T±N), per-task model and agent-preset selection for session tasks, centralized environment-variable management, run logs, per-task concurrency limits and a per-task timeout for script runs.

Shows which graft index a dsh session is serving and whether it still matches the code, and gives the model graft_ask, graft_grep, graft_callers, graft_skeleton and graft_map as native tools.

Session lifecycle: auto-archives sessions older than a configurable threshold (the check interval is configurable too), and an archive panel lists archived sessions for unarchive or deletion; deletion moves the session artifacts to the OS recycle bin, falls back to a plugin-managed quarantine directory where no recycle bin is available, and a ledger re-attaches a session whose artifacts were restored.

Session-affinity keys for LLM gateway requests: it decorates the official pi-ai adapter so Chat Completions calls carry session_id and x-client-request-id headers and Anthropic Messages calls carry a session-derived metadata.user_id, which lets a gateway pin requests per session and raise prompt-cache hits. It registers no provider and takes nothing over — the official llm-pi-ai keeps serving, and disabling or uninstalling the plugin leaves its behaviour unchanged.

Session usage dashboard: day/hour/minute token and request statistics, a 26-week activity heatmap, cache-hit-rate curve, per-model donut and list, and a settings page; it appends its cost item after the official session bottom bar (per-turn cost chips included, no separate detail button), renders the official cache-hit-rate number with two decimals, and estimates costs from multi-condition, time-segmented pricing rules. Re-implemented from HaoyueQin/dsh-usage-statistics-panel with extensions.

Auto-expands streaming thinking: always keeps the latest thinking block expanded, manual actions take priority, only the last block expands when opening a session, and nothing remains after uninstall.

Conversation enhancements: a history-input overlay (Alt+Up, with workspace/session/global/favorite-prompt scopes), recall of interjected messages that were not applied, rewrite-style forking (fork a child session from before a turn and backfill that turn input for editing, including turns whose reply is still running), optional auto-resend after forking (off by default), and a copy-session-id action in the title bar.

Keeps a pool of models and routes every agent step to one of them with its reasoning effort set — configured by clicking in the composer, no JSON needed. The router takes over the composer's model cell, so one control shows the session's model and reasoning effort alongside the router's own settings. Three authorization scopes (model / effort / full) where a manual pick only counts in the scope that owns it, and in effort scope the router is bound to the one model you choose. Task presets with per-model weights, vision routing for image steps, session signals (active todo, context pressure, token budget, subagent depth), and an optional semantic task classifier. Thinking stays on, because DeepSeek rejects a thinking-enabled request whose history contains a tool call made with thinking off.

Interact with branched DSH sessions as live windows on a Loom-style canvas