dsh-capcheck
V0 capability-disclosure scanner for DeepSeek Harness (DSH) cordis plugins: reports which sensitive services (credentials/shell/approval/sandbox/fs/tools...) a plugin declares or references before you trust it.
47 results
V0 capability-disclosure scanner for DeepSeek Harness (DSH) cordis plugins: reports which sensitive services (credentials/shell/approval/sandbox/fs/tools...) a plugin declares or references before you trust it.
Security, Trust & Policy Layer for DeepSeek Harness. Don't trust your agent's tools. Verify them.
深知可信办公全家桶:深知可信咨询、深知可信搜索、深知可信PPT、深知公文写作(skill + MCP 转接),面向 DeepSeek Harness (dsh)
Standalone trusted-LAN Dou Dizhu plugin bundle for DeepSeek Harness
AI coding delivery trust layer for governance-aware coding agents — zcode plugin packaging. The plugin surface lives in skills/ and commands/; this file only provides the npm package identity expected by the zcode plugin host.
ClawShell vision re-implemented as DeepSeek Harness plugins: self-perception (sense), self-adaptation (adapt), swarm trust + ecological niche (swarm), plus tools.
DSH-native multi-AI group chat: the host (user) runs a chatroom of configurable AI agents with per-agent models, role cards, mute controls, and passive/active speaking modes — pure Node.js/Cordis, no Tauri/Rust
DSH 插件安全审查:静态扫描 + 当前会话模型语义审查 + 用户确认
DeepSeek Harness (dsh) LLM adapter backed by aimux — one Rust engine, 325+ providers as dsh model routes.
RTK (Rust Token Killer) native plugin for DSH: auto-rewrites model shell commands to RTK-filtered equivalents through the official tools/execute hook, compressing tool output before it reaches the LLM context.
DeepSeek Harness tools for external opportunity discovery, unmet-need research, pain-point signals, opportunity mapping and validation experiments.
For DSH 0.1.5+, skip the browser-cookie 401 on Hosts already admitted by --trusted-host (Access remains the identity layer) and keep reverse-proxy settings persistence. Not an authentication layer.
JSON 转 Rust:把 JSON 数据转换为 serde Rust 结构体
Indirect prompt-injection guard for DeepSeek Harness: taints tool output by origin and gates privileged tool calls that follow untrusted content
Local tool-call failsafe and DWGR-8 personal governance proxy gate for DSH: blocks high-risk shell patterns, credential leaks, and enforces declarative param-level boundaries with hash-linked JSONL audit logs and optional Gateway.
Click a file path in the chat UI to open it in the running VS Code window owning the matching workspace; Edit/Write rows open a diff of that exact change (webview-safe host route behind the loopback/Origin trust fence).
DSH Settings integration for MiniMax Token Plan usage, using Host webServer routes compatible with trusted bundle plugins and both MiniMax subscription key names.
Unified trusted settings and password access gateway for DeepSeek Harness
Static trust audit for DeepSeek Harness plugins: capabilities, injected prompts, token cost, provenance, and a reproducible 0-100 trust score
Evidence-first inspection, compatibility verification, and quarantine tooling for DeepSeek Harness plugins.
Static linter: every plugin HTTP route must carry a browser-trust fence (loopback Host pin before Origin / sec-fetch-site).
Remote execution worlds for DeepSeek Harness: create a git worktree on a remote machine over a self-hosted node daemon, and run the stock read / write / edit / bash / grep tools inside it.
Remote execution worlds for DeepSeek Harness: create a git worktree on a remote machine over a self-hosted node daemon, and run the stock read / write / edit / bash / grep tools inside it.