Plugins

Leitet Chat-Bilder an ein festes OpenAI-kompatibles Vision-Modell und gibt sachliche Beobachtungen an das gewählte Hauptmodell zurück; sitzungsbezogene Beobachtungen werden über Replays, Kompaktierung und Neustarts hinweg wiederverwendet.

Übersetzungs-Toolkit für DSH-Agents: Satzsegmentierung, Term- und Glossarextraktion, Quell-Ziel-QA (Zahlen, Einheiten, Klammern), Tonfall-Leitfäden und ein dauerhaftes Translation-Memory im Workspace.

Vom Modell verfasste Kontextbereinigung für DeepSeek Harness über die offizielle Compaction-API.

Focus-Board für DeepSeek-Harness-Agents: dauerhafte, vom Modell gepflegte Notizen im Session-Workspace, die Ziel, Constraints und Entscheidungen über Kompaktierung und Sessions hinweg festhalten, mit automatischer Kontextinjektion, Archivierung beim Leeren und einem optionalen Web-Panel.

Pinnt Assistentenantworten über die Web-Aktionsleiste und ruft sie in der nächsten Modellrunde wieder ab (`/pin` `/recall`, mit optionalem Wake).

Bindet das LiuHe-MCP-Code-Toolkit (44 Tools) in dsh web ein: repo_map, Symbolsuche, edit_batch mit toleranter Zuordnung, transaktionale Änderungen mit Undo-Journal und deterministische Qualitäts-Gates – gestützt auf einen absturzisolierten Rust-tree-sitter-Parser-Daemon und einen SQLite-Index.

Werkzeugübergreifendes Sitzungsmigrations-Plugin für DeepSeek Harness (dsh): originalgetreuer Import, Fortsetzung, Wissensextraktion, bidirektionaler Export, Schwärzung sensibler Daten.

Gehosteter Werkzeugserver über MCP: Rechnungs- und Kontoauszug-Extraktion mit arithmetischen Prüfungen im Code, Journalabstimmung in ganzen Cent, PDF- und Tabellenkonvertierung, SQL-Dialekt-Transpilierung und Erreichbarkeitsproben für das chinesische Festland aus Sicht des China-Mobile-Netzes.

Fügt Richtlinienprüfungen für gefährliche Operationen, Ausgabe-Schwärzung und einen Sicherheits-Review-Workflow hinzu.

Lets the model trigger compaction itself at a safe point via a compact_now tool (works mid-turn), plus a context_status tool reporting real token usage against the session's actual context window.

在每条回复下方给出可点的下一步建议，点一下就作为你的消息发出去。Clickable next-step suggestions under every reply; one click sends that sentence as your message.

Execution flow chart: renders the current session's execution as a vertical node flow (turns, user, assistant, tools, approvals, retries, compaction), appended in real time over SSE with auto-follow scrolling — the chart draws wherever the session executes.

Adds a draggable power pill to the DeepSeek Harness Web GUI: a live status dot plus restart and stop actions, where restart is handed to a detached helper that resumes the same boot invocation once the port is released.

Native headless and visible Chromium browser automation agent for DeepSeek Harness with DOM tree extraction, screenshot capture, and built-in SSRF protection.

Fixes the Windows workspace-open action. The shipped opener runs `explorer.exe "<file:// URI>"`, which on some hosts reports success without raising a window — the exit-code verdict cannot tell an opened window from nothing happening. This routes that one action to `explorer.exe /e,<dir>`, which raises a window reliably, and leaves every other application in the menu on the shipped path.

A headless-first macOS Computer Use driver for DeepSeek Harness with AX-first observations, window-bound multimodal evidence, host-owned approval, and evidence-backed action receipts

Drives the GME Test Agent from inside Harness: three tools list interface catalogs, start and repair autonomous test generation in a local Python backend, poll the task until review, and keep PR, skip and cleanup actions behind explicit user consent.

Windows desktop control through UI Automation: nine tools that read a window control tree and drive elements by id through their own UI Automation patterns, with every write gated by an approval policy, an append-only action log and a settings panel.

Tiered compaction hub: replaces the single 80% compaction cliff with a configurable ladder (default 30/50/70/90) whose tiers escalate from gentle to maximum consolidation laws, folds a deterministic keyword index into every checkpoint so facts accumulate across cycles, and summarizes through a model chain — primary plus up to 3 fallbacks, each with its own reasoning effort. Adds a Compaction module sub-tab and a Settings card; toggling off always restores stock compaction.

Context compaction that keeps its receipts: a five-section handoff summary, a durable segment index, review tools that read compacted text back, optional working notes, and a read-only extraction exit. Mounts as the compaction-basic row's replacement; cha

Event-driven Hypatia memory for DSH: automatic conversation logging via session events, queued background consolidation (span summaries + work-unit extraction) with a dedicated model route, and recall injection — zero per-call model work in the main sessi

Host- and agent-layer DeepSeek Harness plugin that prices every automatic context rewrite before it lands, asks the human by default, and reports what the provider had to re-read cold.

Maintained fork of dsh-approval-gate. A neutral operation whose confirmation count has reached the threshold auto-approves when the judge is unavailable, instead of prompting a human again; approver-facing explanations are generated in Chinese from the real sandbox mode, command and paths. Also carries deterministic hard-deny for credential exfiltration and system-path destruction, judge-input redaction, a judge model candidate chain, fingerprint-scoped allow rules, and an approval view with unified diff and one-click revert.

Structured git tools for DeepSeek Harness agents: status, diff, log, branch, commit and stash. Output is pinned with ten `git -c` overrides so a user pager, colour or quotepath setting cannot reshape what the agent reads; every invocation is an argv array, so no shell ever parses an argument; and the mutating actions go through the harness approval service, refusing rather than proceeding when no approval service is mounted.