Plugins

Desktop-Pet für die DSH Web UI mit 25 transparenten Animationen, Umherwandern auf dem Bildschirm, Klick-Reaktionen und Ziehen, plus einer reproduzierbaren Asset-Generierungs-Pipeline.

Quantitatives F&E-Toolkit für DeepSeek Harness — 46 Tools in sechs Bereichen, die Marktdaten, Indikatoren, Faktorbewertung, Walk-Forward-ML-Validierung, Risiko, Optionen, Anleihen und Fondssimulation abdecken, mit einer End-to-End-Research-Pipeline.

Hochverfügbarkeits-Failover für Modelle (Quarantäne, Circuit Breaking, Wiederherstellung per Probe) und Subagent-Orchestrierung (fanout/pipeline/supervisor) mit zweisprachiger Einstellungs-UI.

Portiert alle 34 offiziellen opencode-Themes (33 statische JSON + 1 System) in DeepSeek Harness: datengesteuerte Pipeline (JSON → Farbauflösung → DSH-Override), browserabhängige Persistenz, Wahl zwischen Monospace/Proportional, fünf Code-Schriftarten und ein System-Fallback, der nur die Typografie beibehält.

Bindet DeepSeek Harness in eine Hermes-Pipeline ein: Dispatch-Spec-Vorlage, Modell-Tier-Routing, Qualitäts-Gates für Orchestrator-Läufe, Git-Single-Writer-Regel, als SKILL.md-Paket (als Bundle installierbar).

Virtuelles Produktteam-Agenten-Preset: Die benutzergesteuerte Konversation führt Sie durch die vollständige 12-Schritte-Pipeline von der Idee bis zum ausgelieferten, beworbenen und betriebenen Produkt, mit stufenweisen Gates und stufenspezifischen Skills.

Plugin trust pipeline for DeepSeek Harness: deterministic static scan with verdicts, opt-in runtime guard with honeypot lures, agent audit-protocol skill, and a browser shield status light. Alarm-only, never an enforcer.

Contract regression harness for plugin authors: a scripted LLM adapter drives the real pipeline in isolated child processes with session-event assertions (load/contract/behavior/fault layers); zero API key, CI-ready.

Runtime security gate on the tool pipeline: denies calls naming hosts outside an egress allowlist, redacts credentials from results at the canonical value rather than only the rendered content, and appends every decision to a JSONL audit log; ships in monitor-only mode.

WeChat Official Account content studio: anti-homogenization rotation writing, low-creativity remediation playbook, blessing-image visual baseline, gpt-image pipeline with OCR acceptance, and the measured xiaolvshu draft web API.

Layered memory pipeline for DSH: auto-distills conversations into atomic facts, scene summaries and a persona profile (L0–L3), with hybrid BM25 + vector retrieval, chat/work family separation, and context injection before each model step.

Declarative multi-agent pipeline executor for DeepSeek Harness (dsh): runs a manifest of confined role dispatches with mechanical gates and a machine-written evidence ledger. dsh 声明式多角色流水线执行器：按 manifest 派发受限子代理、机械门禁、机器写证据账本。

Model-agnostic visual-inspection pipeline for text-only agents: page-by-page HTML screenshots plus a ready-made vision-subagent briefing contract (gewu_prep), then source-code truth verification of every finding (gewu_locate); validated on mimo-v2.5 & qwen3.7-plus.

Cross-session reference (mention) for DeepSeek Harness: paste @[label](dsh-session:…) into any session — including one in another workspace — and the host resolves and injects the referenced session snapshot via the native session-reference pipeline. Incl

Multi-agent deep long-form article pipeline skill for DeepSeek Harness, running 8 roles across 5 phases with parallel literature/data/case retrieval, a critical-companion review, M-Gate mechanical final check, independent G0-G13 audit and human verification gates.

Rigorous open mathematics research suite: four agent skills (rigorous-open-math-research, manage-math-research-program, math-research-workflow, lean-verify) for theorem solving with adversarial audit, research program management, pipeline orchestration, and Lean 4 formalization audit; CI-verified tests and mechanical upstream sync.