Plugins

Project-level memory built around a ledger whose structure the AI designs per project; the plugin only handles detection, injection, write-back, archiving and git. Memories are split by meaning: the active ledger holds current state, the turn journal is append-only, and decisions, the goal tree and lessons are one-card-per-file (with a multi-dimension index over the card folders); the fundamental principles file can be changed only by the user. There is a single compaction mode: before compacting, the AI sorts the work into three buckets - what must stay in context and is re-injected verbatim, what is written to the project's _stash/ for later lookup, and what is dropped. When a conversation is opened in a container workspace holding several projects, it can be locked to one sub-project (the lock lives in the plugin's own home and survives restarts; it leaves no marker in your project). Also an activity log with five levels, where L4 stores the user's own words verbatim (never injected, queryable on demand). Before each compaction a verbatim capsule (user quotes, raw errors, changed files) is written to disk, and both the capsule and its source material survive a restart; a handoff document lets a new conversation resume the previous one. 21 tools, 12 subcommands, 39 settings (one of them switches whether the settings panel or each project's manifest wins). Runs locally, no runtime dependencies.

Manage skills and MCP servers from one DSH settings panel. Skills are read from disk: browse, search, preview the body, create, rename and delete, with a backup written before every removal. Global MCP servers are the real cordis.patch.yml loader rows; per-project entries mount only for sessions in that project. Each expanded card lists the tools that server actually registered, and a per-server allow/deny rule keeps only the ones you need — a blocked tool never enters the tool registry, so it costs no context on every turn. Every MCP card carries a connection status tag (connected with a tool count, retrying, gave up, activation failed) plus a one-click reconnect, so a server that fails at startup is visible instead of silently missing its tools. Imports existing configs from 26 other MCP clients.

核心：信封／签名／握手／幂等／落盘 · Core: envelope, signing, handshake, idempotence, on-disk store

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).

Per-project memory for DeepSeek Harness. Three Markdown files per workspace — PROJECT.md, DECISIONS.md and SESSIONS.md — are auto-created on the first step of any session in that workspace, re-read from disk and re-injected only when they change (and re-injected if compaction dropped them), and written back by classification through a memory_checkpoint tool, with a bounded end-of-turn nudge when a turn did work but recorded nothing. memory_read and memory_search read the live log and the archive, the latter over a zero-dependency BM25 index. The session log is capped at 200 entries with the overflow moved to SESSIONS-archive.md — never injected, but visible, readable and searchable, and restorable one entry at a time. Settings page: a workspace list with a path filter, read-only tabs for the three memory files plus the archive, reload, export/import of one workspace's memory as a JSON bundle, a two-step clear, a panel that rewrites or removes the AGENTS.md instruction block, and a PROJECT.md staleness reminder. A composer status chip shows that workspace's sync state and creates the files on click. Zero dependencies — no vector store, no embeddings, and no model calls.

Windows system-drive (C:) cleanup for DSH: rule-based scan, five-tier safety grading, a staging area with a rollback ledger, UAC elevation, and migration to another drive — a model tool plus a Web GUI panel.

Electrical and electronics calculation plugin for DeepSeek Harness: every computation runs through dedicated tools on self-describing complex value objects in SI base units, with an isolated calculation agent preset and a records panel that archives every solve on disk.

DSH web plugin: open .docx/.xlsx/.pptx in the better-sidebar editor through a self-hosted ONLYOFFICE Document Server (JWT-signed config, in-network document/callback routes, save-back to disk).

Sidebar session-title icons: an auxiliary model call on the same route as title generation draws a 24x24 monochrome SVG metaphor shown before each session row title, with disk caching.

Session export and compliance archiving for DeepSeek Harness: deterministic redaction, Markdown/JSONL/PDF outputs, on-disk archive management, and a trend/audit view.

DSH workspace hub: organize workspaces into color-coded folders in the sidebar (three-level drag reorder for folders/workspaces/sessions, inline edit, one-click expand/collapse, dual-channel search) with per-workspace and per-session token usage and cost stats from real session logs (official DeepSeek peak/off-peak pricing for any deepseek model, three spend calibers, hover detail cards, overview panel) plus OpenCode Go plan usage; data persists to host disk and refreshes live every 30 s.

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.

System status floating widget: live CPU/memory/disk usage in the bottom-right corner of the DSH Web UI with threshold colour warnings.

Incremental file mounting with read dedupe: mounted line ranges are never re-sent to the model, on-disk changes invalidate and remount, with a Mounted Files tab and token-savings accounting.

DeepSeek Harness plugin for rulemux: asks the local rulemux CLI for the rules rendered for this workspace (rulemux inject --agent dsh) and injects the output into the session at the first user turn, once per session. Nothing is written into the workspace for rules; skills are the only part that lands on disk.

MCP gateway for dsh that connects each server on first use and disconnects it after an idle window; tool lists are cached on disk, existing mcpServers files can be imported, and child-process environment variables are scrubbed.

Disk-backed long-term memory for DSH: a SQLite FTS5 vault that captures every conversation turn, injects relevant past turns into new sessions within a per-turn character budget, groups sessions into resumable conversation files, and ships a web panel for search, paging and capacity.

Adds a red "Delete session" item beneath "Archive" in the sidebar session menu, with a custom confirmation dialog, for real session deletion — stops the session's running work, detaches its workspace, archive and pin accounting, removes its projection cache and log directory, then broadcasts removal so the row disappears at once, instead of only archiving it. Session deletion removes the conversation from disk permanently and cannot be undone.

Per-request token ledger for DSH with four-bucket accounting, a 53-week heat map and a live concurrency gauge, plus an on-disk JSONL ledger that outlives the 600-row session projection window and is swept back from every stored session log at startup.

System monitor for kids: a single card summary of CPU, memory and disk.

Storage explorer: disk boxes, the home folder and the heaviest items.

Computer safety shown as castle guards: firewall, disk encryption, lock screen, remote ports and login history.

Scheduled update checks for DeepSeek Harness itself and installed plugins, applied only in idle windows — four schedule modes (daily, every N hours/days/minutes), bidirectional plugin whitelist, disk-verified installs, and a web settings panel with hot-reload config.

Adds a Forget action to every session row menu, confirmed by a second click in the menu, that removes the session from the sidebar and from disk: running work is stopped, workspace, archive and pin records are detached through official channels, the log directory and projection-cache entries are deleted, and a revive guard re-deletes anything a late write or a lingering handle recreates; memories, attachments and other sessions are left alone.