Plugins
Browse, filter, and install DeepSeek-Harness plugins.
140 plugins found
dsh-stability-audit
chunfenxiazhi-collab/dsh-stability-audit
Stability audit for installed dsh plugins: static risk grading (hook surface, startup work, inject, entry, dep ranges) plus optional isolated install verification.
dsh-feishu-channel
wyattjhayes/dsh-feishu-channel
Security-focused Feishu (Lark) channel for DeepSeek Harness: allowlisted remote-agent access with workspace-scoped paths, symlink checks, risk-based approvals, session isolation, redacted logs, and bounded message queues.
weiwen-law-dsh
shaky77/weiwen-law-dsh
White-box causal guardrail for dsh: every tool call is adjudicated before execution along a deterministic causal logic chain — blocks destructive and credential-file operations, escalates repeated boundary violations, cuts faulting links until verified-fixed, and gives risk verdicts on un-auditable execution; runs fully local with zero API cost, plus 6 white-box self-check tools.
dsh-popper
1473382/dsh-popper
Falsification-driven correction loop for agent sessions: risky work commits an evidence-checkable claim first, deterministic gates verify it, falsified claims force mutually exclusive replacement hypotheses with discriminating experiments, and every event lands in an append-only evidence ledger.
riskproof
onlyqzq/riskproof
Provenance-aware execution security for DeepSeek Harness. Track sensitive data across tool calls and stop risky side effects before execution.
dsh-codex-approval
040822/dsh-codex-approval
Codex-style approval autopilot for DeepSeek Harness: ordered glob rules (allow/ask/deny) plus an AI risk judge (low/medium/high) mapped through a risk tolerance, as an approval answerer.
safety-net
johnxu22786/safety-net
Destructive-command interception gate for dsh: parses shell semantics, judges risk against 41 built-in rules, and holds irreversible rm -rf, git reset --hard, and git push --force style commands at a confirmation gate.
dsh-update-copilot
hezhongtang/dsh-update-copilot
Update copilot for DeepSeek Harness: one scan covers the DeepSeek Harness core and every profile plugin; explains what changed and how risky each update is, then updates only what you confirm.
dsh-plugin-clinic
ayahunter/dsh-plugin-clinic
Read-only health clinic for the installed DSH plugin set: loader health, dependency integrity, version compatibility, install-script risk, duplicates and patch integrity, delivered as a model tool, a Settings dashboard and JSON reports.
agent-loop-workflow
lesliewylie/agent-loop-workflow
A project-agnostic multi-agent collaboration protocol for the DeepSeek Harness: loop guards that stop runaway agent loops, a six-field handoff format, risk-tiered review routing, and a fixed verify→commit→push→review→close sequence. One skill instead of the same rules copy-pasted into every agent's instructions.
DSH-guardrails (dsh-guardrails)
vladimirkuzen/dsh-guardrails
Prompts or refuses risky tool calls before DeepSeek Harness runs them, and keeps a local hash-chained audit trail; Windows-only, zero runtime dependencies.
dsh-contract-redline
liuhange789/dsh-contract-redline
Contract risk quick-scan for small businesses: paste a procurement / supply, lease or service contract and get a checklist of high-risk clauses, unbalanced obligations and missing key clauses, each with a short excerpt and a plain-language reason. Rule-based and local-only; no OCR, no network, no legal advice — it never says whether to sign or whether the contract is valid.
dsh-vertical-kit (dsh-weekly-report-check)
zjukop3/dsh-vertical-kit
Weekly-report hygiene from markdown: what was done separated from what is planned, risks and blockers named, and at least one quantified item so progress is tellable from motion. Its HTTP routes have no DSH access check, so a web page open in your browser could call them.
dsh-modpack-kit
vesna-strivozha-cn/dsh-modpack-kit
Pack several DeepSeek Harness plugins into one .dshpack and install them in one pass. Both formats ship a standalone installer, so a machine with only a bare DSH can install (payloads go to a local vendor folder, installed plugins are skipped unless --replace). Packing always scans for secrets and risky code; high-risk hits pause until you confirm, then the same job resumes without scanning again. Installing shares the progress dock, skips plugins you already have and then lets you replace them with the packed versions; when you are online the market version is preferred. Packs can be opened with a file or folder picker, any directory, an imported-packs list or a link. Categories, search, paging and same-name entries come from the plugin-market catalog.
copyright-risk-detector
onion716/copyright-risk-detector
小红书虚拟商品违规风险检测工作台:对商品名称、标题、正文、配图与交付文件做本地规则筛查与 DeepSeek 复核,输出风险分级、命中明细、改写建议与运营处置建议。
dsh-jev-auto-approve
vultwo/dsh-jev-auto-approve
Reviews DeepSeek Harness approval requests with TypeSafe Jev and auto-approves low-risk, explicitly authorized actions; includes an English settings page for the key and token/cost usage. Each approval request is sent to the third-party TypeSafe Jev service (api.typesafe.ai) with your key, and low-risk ones are approved automatically.
dsh-trading-workbench
clusm/dsh-trading-workbench
Trading workbench for DeepSeek Harness: an AI decision chain with a per-day round archive, Demo execution and risk views, backed by a local Python trading service that bootstraps itself on first run. GitHub v0.1.1+ checks every HTTP route through DSH request admission; npm 0.1.0 does not include this fix.
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-plugins (tool-policy)
genesi599/dsh-plugins
Blocks risky tool-call shapes before execution: inline multi-line Python, forwarded content without a source, missing source labels, and memory writes past the character cap. It also asks for confirmation before shell commands, and one rule is specific to the author's WeChat sending workflow.
dsh-agy
yhan-sun/dsh-agy
Brings Google Antigravity (agy CLI) models into DSH with streaming, thinking and native tool cards, a manual multi-account pool with quota and proxy controls, and an optional experimental direct transport. Community fork of amlyczz/dsh-agy-link. Google's Antigravity FAQ says personal plans do not support wrapping agy in third-party coding agents, so using it may put your Google account at risk of restriction. Needs Node.js 24 or newer and Google's agy CLI on PATH; the experimental direct mode is off by default.
dsh-approval-why
zhknb/dsh-approval-why
Adds a “why does this need escalation?” button to every DSH approval card: it opens a panel with the permission requested, the risk, what refusing means, the exact tool arguments and whether the named paths fall outside the session workspace; the card also keeps a one-line risk and refusal consequence visible under it, and the model’s original reason is split into labelled paragraphs instead of a run-on block.
dsh-slowmist-security-guard
wjl906/dsh-slowmist-security-guard
Security guard for DeepSeek Harness (Cordis): reviews every tool call in pre-execute against SlowMist agent-security rules, denying or flagging risky commands, MCP servers and skill loads, external URLs and repositories, untrusted documents, and EVM or Solana addresses; scans tool results for prompt injection after execution and ships a slowmist_scan audit tool plus an injected security policy prompt. Rules follow the open-source SlowMist Agent Security Skill (MIT).
dsh-server-connector
niechao136/dsh-server-connector
DSH plugin: SSH execution for servers configured on this machine, with a built-in high-risk command policy (block / confirm / allow).
dsh-ps-preflight
dingchenhui0618-arch/dsh-ps-preflight
Grades every PowerShell command an agent is about to run before it runs: deletion, overwrite, network download and credential access are four required classes, plus dynamic execution, system and privilege changes. Verdicts are a pure deterministic rule function — no model call, no network, sub-millisecond — with two escalation rules (download+execute, credential+exfiltrate) that jump straight to high. High risk stops the tool call for a second confirmation (in-panel queue or the native approval), deny or timeout means it never runs; medium is logged but never blocked. Every command and its verdict, including the redacted command text, is appended to a local JSONL audit log. Quoted danger strings are treated as data, -WhatIf is respected, -EncodedCommand is decoded and analysed again. Host + client halves, zero runtime dependencies, 55 rules, 136 tests.