DeepSeek Harness 플러그인 — 748 페이지

J-Space Anchored Warm (j-a-w) agent presets for DeepSeek Harness (dsh): two-phase anchored bootstrap + warm-up round + J-Space protocol injection, in two tool-surface variants — j-a-w(Pwsh) (Windows) and j-a-w(Bash) (cross-platform). Ships as an installable bundle that syncs the presets into ~/.dsh/.agent-presets on host startup.

牙买加国家

Deterministic artifact lifecycle for DeepSeek Harness: scans plan docs and scratch scripts left behind by agent sessions, returns a reviewable cleanup plan — the janitor is never an agent

日本国家

SQL 优化模式：EXPLAIN 分析、索引策略、N+1 解决、查询改写。受 wshobson/agents（38k★ MIT）启发。

Golden Spatula fan theme for the DeepSeek Harness Web GUI: an original hextech arena backdrop with a golden spatula crest, four accent palettes, a hex-board lattice, aura/dust ambience, soft or sharp corners, and a floating hex-core panel. 金铲铲同人主题：原创海克斯竞技场背景、四套强化色、氛围光效、圆润/锐利棱角与海克斯核心面板。

Jenkins CLI plugin (dual-face): multi-server config page, model tools, and a per-workspace Run-Jenkins-Job entry with environment tabs. Bilingual UI (zh/en), no hardcoded paths, publishable to npm / GitHub

Jenkins integration for DeepSeek Harness: a live build-progress panel in the Web GUI plus model-facing tools for triggering builds, reading stages and console logs, and browsing a build's workspace

时差调整

TypeSafe Jev (System One) typed-decision tools for DeepSeek Harness (dsh): ask noul/choice/score questions and get structured answers instead of prose

TypeSafe Jev System One semantic tool routing and typed decisions for DeepSeek Harness (DSH)

Ask Jev (TypeSafe System One) typed questions from DeepSeek Harness: batch judgements with probabilities, configured per user in Settings.

Use the Jev (System One) decision-model paradigm with any OpenAI-compatible LLM — no TypeSafe key required. Registers a jev_decide tool on DeepSeek Harness (dsh).

Jev (TypeSafe System One) gives the human a second opinion, not the agent. When DSH's model asks you a multiple-choice question, dsh-jev-advisor builds a structured Jev request and floats the recommendation, probability spread and confidence beside the options - one click to adopt. API key lives in Settings. No prompt changes.

Data-driven preflight and evidence-aware context gates for DeepSeek Harness

Search-flood adjudication for DeepSeek Harness: a jev_search tool that runs ripgrep and pipes every hit through TypeSafe Jev's typed judgments (noul/choice) so the model reads ~30 calibrated rows instead of hundreds; a flood nudge on grep results; provider url/key/model fully configurable; degrades to plain grep behavior whenever the endpoint is unreachable. Host-half only, zero npm dependencies, cross-platform (Win/mac/Linux).

DSH 的执行前安全阀门:用 TypeSafe Jev 判定 shell 命令/脚本是否会不可逆地删除或覆盖真实数据。四态(allow/revise/block/escalate),挂在 tools/pre-execute 上。零依赖;支持 WSL/Linux(bash)与 Windows(pwsh);含一次性令牌、额度耗尽降级、共享审计日志。

A judgment kernel for coding agents: bounded questions a small Jev judge answers, so the main model keeps its context and tokens for real reasoning.

Jev (TypeSafe System One) judgments at the DeepSeek Harness agent-loop gates: pre-step, pre-execute, post-execute, and turn-stopping. No UI dependencies; mount it in any Harness composition.

Typed, auditable long-term memory for the DeepSeek Harness: a turn-end write hook, a judgement layer (Jev or a deterministic fallback), a local store, and per-session recall injection.

Ask Jev (TypeSafe System One) typed questions from DeepSeek Harness: batch judgements with probabilities, configured per user in Settings.

Jev 决策模型路由插件：默认只调思考强度（缓存安全），可选自动模型路由（默认关闭 + 缓存代价硬门禁）· DeepSeek Harness plugin

Cut LLM costs: route routine tasks to cheap subagent models (DeepSeek Harness plugin)

Drive a browser from one natural-language goal: TypeSafe Jev picks the next operation and target from an indexed element table, so a DeepSeek Harness session spends one tool call instead of one turn per click.