Plugin

Role-based multi-agent team for DeepSeek Harness. The /team command assembles up to 12 specialist subagents (product, architect, researcher, UI, backend, frontend, database, security, reviewer, QA, DevOps, docs) and runs a nine-phase gated pipeline (clarify, research, design, spec review, plan approval, implement, review, test, deliver) over a shared workspace of artifacts (SPEC, PLAN, TASKS, STATE, REVIEW, TEST). Quality gates are enforced by plugin code rather than requested in prompts: an unfinished task cannot be marked completed, quality findings must be adjudicated by the reviewer or QA role, and coverage gaps or rework over budget appear live in the overlay and in /team status. Ships a live team overlay (phases, roster with per-role models, task DAG, artifact preview, decision buttons), a persistent-team mode that resumes across sessions, and per-run token and time accounting. Zero runtime dependencies, no build step.

会话记忆 → OKF 知识沉淀插件:预测驱动的神经自我学习记忆系统。把会话中高价值内容按 OKF v0.1 规范沉淀为长期记忆,跨会话自动唤起。Session-to-OKF memory plugin with neuro-self-learning (predictive recall, uncertainty-driven capture, reinforcement feedback).

Unit conversion toolbox covering 20 categories — length, mass, temperature, area, volume (incl. US cooking units), speed, time, data sizes (decimal MB vs binary MiB), data transfer rates (Mbps vs MB/s), acceleration incl. g-force, illumination (lux / foot-candle), pressure, energy, angle, frequency, power incl. three horsepower definitions, force, torque, typography (px/pt/em/rem), and fuel economy (mpg ↔ L/100km). Two tools — convert_unit and list_units — zero runtime dependencies, pure arithmetic; each result includes the formula applied.

Guardia di disciplina dell'agente per DeepSeek Harness: quattro cancelli applicati al confine dell'harness — interruttore di loop, fusibile del costo dei token, avviso di cambio rotta, approvazione del piano — dove i promemoria a livello di prompt non arrivano, così un agente non può più bruciare una quota settimanale in pochi minuti; chip 🛡 con un clic, soglie ricaricabili a caldo e preset da invisibile-ma-corazzato a solo avviso.

Strumenti Odoo in sola lettura su JSON-RPC: info sul server, introspezione dei campi dei modelli e una search_read limitata a una allow list di modelli i cui nomi dei campi domain non possono contenere punti, così la navigazione verso record correlati non è possibile. Uno strumento di creazione bozza viene registrato solo quando allowWrite è abilitato, ed è limitato a sale.order e project.task con lo stato di bozza imposto dal plugin.

Aggiunge controlli di sessione, progetto e globali nell'interfaccia web DSH per competenze, server MCP, strumenti, iniezioni di prompt, escalation delle approvazioni e guardrail di sicurezza; l'applicazione è limitata all'agente corrente.

Gate di verifica per gli agenti DSH: ogni criterio di accettazione deve essere supportato da prove reali degli strumenti timbrate dal server prima che il gate di completamento lasci passare un obiettivo (audit consultivo, gate di applicazione, permessi durevoli).

Applicazione dei conventional commit, generazione del changelog, riepiloghi PR e push dei branch tramite la semplice CLI git.

Disciplina di sviluppo spec-driven per dsh: cinque skill keel (anchor/spec/probe/build/audit), tre strumenti di revisione e sei template di spec impongono le specifiche prima del codice, la verifica delle assunzioni e il controllo dello scope.

Plugin di imposizione del protocollo memoria: collega il server MCP opencode-memory, forza memory_weave prima delle chiamate agli strumenti, acquisisce automaticamente i turni e inietta il contesto della memoria nei passaggi dell’agente.

Model-callable compact_agents tool that force-compacts context in every live session (main session, ordinary sub-agents and AgentTeams members alike), queues busy targets and compacts them when the turn ends, reports the shadowed node count and estimated tokens per target, and ships a browser form for the trigger ratio, retained ratio, controlled-phase output budget, auto-continue budget and an end-of-turn pre-compaction ratio that compacts near the trigger line once a turn ends, so a reply no longer opens with a wait for the summarizer.

DeepSeek Harness plugin for Agent-Playbook: multi-agent claims, isolated git worktrees, and enforced acceptance checks. "Done" is an exit code, not a claim.

Connects DeepSeek Harness to a local SiYuan (思源笔记) desktop app: registers its note tools as mcp__siyuan__<tool>, exposes the AI configured in SiYuan together with its agent loop as one more tool, ships a notes skill, and enforces a readonly/authoring/full profile on every call while keeping the tool list available when the app is closed.

Team cost captain: meters every model call (subagents included), attributes each call to a member by workspace rules, prices it with the built-in DeepSeek 2026-09 catalog (peak/off-peak, CNY/USD) or user overrides, and appends it to a local JSONL ledger. Adds hot-reloadable member budgets, 80%/100% webhook alerts, optional over-budget enforcement, a team_cost_report tool with CSV export, and a settings-page team dashboard served over a Typert RPC service.

Open folders on remote hosts over SSH as Harness workspaces: file tools (read/write/edit/grep/glob) and shell commands run directly on the remote host over SFTP, ripgrep, and ssh2 exec, with no local mirroring, an encrypted multi-host registry, and the local sandbox policy still enforced.

AST-based context compression, test log filtering, and token budget guard for DeepSeek Harness: strips redundant syntax, prunes noisy log lines, and enforces a token ceiling so the context window never overflows.

Three independent language channels for model output — tool-call descriptions (every tool-call card title, including run_code in PTC-style presets), thinking, and replies — each set to follow the web GUI language, force a specific language, or turn off (default: only descriptions, following the GUI). One global runtime-context prompt directive on the host; the client half reports the GUI language; a Settings plugin card switches the modes.

Forces periodic Engram memory recall and save reminders for DeepSeek Harness agents: injects a dynamic system-prompt context that reminds the model to load prior session memory at the start of a session, and to save new decisions/bugfixes/learnings every

Single-password authentication gate for the dsh web surface: configurable session validity, Ctrl+Shift+L lock, CLI password management, brute-force lockout, in-UI password controls.

Branch-role guard for AI coding agents: configurable integration/preview/production/archive roles, blocking direct push, force-push and delete on protected branches and agent merges into production/archive.

Dual-model auto-routing plugin: low-cost direct / high-cost upgrade with an escape-learning closed loop (wrong direct answers auto-learn fingerprints, force-upgrading the same fingerprint next time), persisted and interoperable with the Python ModelRouter.

Safe, audited SQLite/PostgreSQL/MySQL access for dsh agents: schema introspection, enforced read-only queries, a write approval gate, and a durable JSONL SQL audit trail.

Enforce configurable git commit / push / PR conventions for DeepSeek Harness agents

DSH sidebar footer ordering: forces the sidebar.footer.action slot into a vertical stack and lets users configure the top-to-bottom order of its entries.