プラグイン
DeepSeek-Harness プラグインを閲覧・絞り込み・インストールできます。
142 件のプラグインが見つかりました
agent-clinic
1687210989/agent-clinic
DSH session log の読み取り専用診断:22種類の名前付き診断(token浪費、retry/失敗ループ、context 膨張、session 間の記憶喪失、idle MCP server/skill)。各指摘には `session.v4.jsonl.zstd#L101-L110` のような再現可能な証拠 locator と、実行可能な処方を付けます。処方は profile-patch YAML または現状→提案値を示す command で、risk、verification、source を含みます。却下する処方を feedback command で廃止でき、報告書は feedback で除外されたと明記します。依存ゼロ、既定で外部通信ゼロ、session file は一切書き換えません。
dsh-ocr-review
jayyuen666/dsh-ocr-review
外部ocr(open-code-review)CLIを7つのモデルツールから利用。review、全ツリー走査、決定論的なファイル選択/ルール雛形、コメント絞込み付き履歴、比較/書出し、ルール診断、読取専用LLMカタログ照会を備え、形式/型/列挙値を検査。OCRの検出結果と選択ファイルの確認範囲は別々に報告します。設定ではホストカタログからprovider/modelを選び、CLIのmcp_servers JSONを編集、ocr auth状態確認、ChatGPTモデル切替、CLIバージョン表示が可能。外部CLIは別途必要で、プラグインは導入/更新しません。
dsh-composition-doctor
lemonxiny55/dsh-composition-doctor
対応対象のDSHプラグイン障害について、合成元のパスと手動の次手順を説明します(重複挿入、bundleパッチ欠落、peer不整合など)。独立CLIとレポートビューアーを備え、既定ではオフラインの読み取り専用診断を行います。観測していない実行時状態は不明のまま扱います。
dsh-agent-quality-diagnosis
cat552/dsh-agent-quality-diagnosis
DSH agentセッションの実行可能な品質診断。tool callの証拠と次の手順の提案を示します。
dsh-accessibility
omdsh-dev/dsh-accessibility
DeepSeek Harness向けscreen reader案内とアプリ内accessibility診断。
agents-plugins
lxp731/agents-plugins
DSHサービス制御:`dsh --profile ctl`からsystemdのサービスライフサイクル、設定、診断を管理。
dsh-healthcheck
taxueseek/dsh-healthcheck
読み取り専用の DSH 環境ヘルスチェック:ディスク、メモリ、レイテンシ、~/.dsh の肥大化、古いプラグインの検出を、段階的なレポートとベースラインのトレンド比較付きで実行。何も削除しません。
dsh-honcho-memory
kingcharleslzy-ai/dsh-honcho-memory
DSH プラグイン:自動デュアルピア Honcho v3 メモリ、コンテキスト呼び出し、検索、推論、結論、ヘルス診断。
dsh-code-check
a179-sanae/dsh-code-check
自動型チェック診断:モデルが TS ファイルを編集・作成した後、バックグラウンドで `tsc --noEmit` を実行し(プロジェクト単位でデバウンス)、`code_check` ツールでエラーを報告します — ファイル別に行情報・重要度・TS エラーコードをまとめ、「編集→チェック→修正」のループを閉じます。
dsh-ci
xxiaoxiong/dsh-ci
DeepSeek Harness のプロバイダー非依存な CI/CD 機能。GitHub Actions 対応、範囲を限定した失敗証拠、ソースを意識した診断、承認を要する再実行/キャンセル操作。
dsh-plugin-socratic-tutor
bortpro/dsh-plugin-socratic-tutor
Socratic tutor with topic-specific curricula, diagnostic questions, a saved skill map, spaced review, and a final assessment.
dsh-plugin-profiler
liuhange789/dsh-plugin-profiler
Plugin load audit for your installed set. `/diet` prints one card: context tokens per request, hot-path host time outside the tool body, listener fan-out per plugin, and which three plugins to look at first, plus per-tool context pricing. Reads the running event bus and tool registry only, never patches another plugin, writes no files and makes no network calls. The card states its boundary: rc.7 exposes no per-listener timing, so the cut list is fan-out, not per-plugin milliseconds.
dsh-notify-context
leerolling/dsh-notify-context
Desktop notifications for turn completion, pending approval and pending questions, each with a per-situation title and the task summary it carries; silent while the client window is in front, and backed by a rotating diagnostic trace. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-healthkit
zuoyunlai/dsh-healthkit
Read-only health check for DeepSeek Harness: scores the install 0–10 across seven weighted dimensions, lists findings with fix hints, and verifies whether previously reported findings were actually fixed. Ships 16 tools and a native sidebar panel. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-mode-router
guyeqiuchen/dsh-mode-router
Classifies each incoming message as chat or work by rule-based scoring and injects the matching answer protocol, with manual /chat, /work and /auto overrides and a mode_status diagnostic tool.
dsh-connect-loomy
diudiu62/dsh-connect-loomy
Registers the Loomy desktop app's locally signed-in chat models as a `loomy` provider in DSH. Several accounts can be signed in at once and switched from a drawer on the composer badge, with each account's points shown per row, optional per-call rotation between healthy accounts, a settings section with per-model toggles and an upstream diagnostics table, and a status/doctor CLI. Shares the `loomy` provider id with 2239058652/dsh-loomy-connect, so the two cannot be installed together; this one differs by the account drawer, the per-account points breakdown and the CLI.
dsh-r-runner
seally-web/dsh-r-runner
R in the sidebar: an RStudio-shaped workbench backed by a persistent R session over a loopback driver (6-9 s cold boot, then 12-84 ms per run). R syntax highlighting with a multi-buffer editor, file tree with extra folder mounts, environment pane you can edit in place, plot gallery with a floating viewer (zoom, resize, open original), five bottom tabs (console with prompt, persistent history, ?topic help, sortable data viewer, package manager with install.packages), completion / diagnostics / go-to-definition, .Rmd chunk runs, Rmd, Quarto and Shiny previews, and an agent-facing r_run tool. Requires a local R installation. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
deepseek-harness-plugins (dsh-plugin-doctor)
codenestra/deepseek-harness-plugins
Plugin health diagnostics, conflict detection, dependency compatibility, installation history, and audited disable/restore actions.
deepseek-harness-plugins (dsh-turn-profiler)
codenestra/deepseek-harness-plugins
Per-turn latency waterfalls, live timing, token and prompt-cache diagnostics, and JSON export.
dsh-experience-optimizer (plugin-trim)
starlit-sky-code/dsh-experience-optimizer
Plugin diet and compatibility diagnostics for DeepSeek Harness: scans context cost and plugin/runtime/profile conflicts, then applies low-risk reversible trims only after confirmation; ships a Web settings console.
dsh-latex
yaohua-leo/dsh-latex
LaTeX workspace for DeepSeek Harness: opening a .tex file shows a line-numbered editor in the right sidebar, saving writes the source and builds it with the local TeX Live latexmk, and an open PDF preview refreshes in place. Includes structured file:line diagnostics, SyncTeX, last-good preview retention and seven agent tools.
dsh-steward
stallenlu/dsh-steward
Read-only diagnostics for a DSH install — boot-crash logs and the cause named in them, config drift against a recorded fingerprint, backup freshness, workspace growth, plugin updates, host-version compatibility, duplicate client ids, and model availability — plus an action layer that always shows a plan first and needs a single-use fingerprint and explicit confirmation before anything is written. A checkup also runs 45 seconds after every host start; when the 3-hour probe cache is stale it sends a one-line test prompt to the default model and a few models of each configured provider (model.probeOnReport can switch this off). Registers the agent tools dsh_steward_report and dsh_steward_act (clean caches via dsh-disk-cleaner, backup via dsh-config-manager, disable or enable a bundle, set the default model; execute takes the plan's hash plus confirm: true) and a system-prompt section telling the agent to run the report first when something crashed.
dsh-task-progress
himleoing/dsh-task-progress
A draggable task progress panel with local estimates, independent teammate progress, optional AI observations, and probe cost diagnostics. Release candidate for the Web profile only; needs Node 24.9 or later and Harness >=0.2.2-rc.1, tested on 0.2.2-rc.2. AI observations are off by default and, once enabled, spend tokens of the session's provider. It uses the same package name and mount id (dsh-task-progress) as chen8923/dsh-task-progress, so install only one of them.
dsh-toolbox
hao0311/dsh-toolbox
Toolbox panel for the DeepSeek Harness Web GUI: a right-side drawer (full screen on phones) with connection information (Wi-Fi details and gateway ping need a Windows host), one-click diagnostics, a speed test with server-side timing, a quality monitor, history and auto patrol; the host registers the /api/toolbox routes.