dsh-harness-audit
Audit an agent harness against the harness-evaluation criteria, with machine-enforced evidence validation.
71 results
Audit an agent harness against the harness-evaluation criteria, with machine-enforced evidence validation.
NeevSandbox execution providers for DeepSeek Harness
Installation safety gate & data-protection guard for DeepSeek Harness: 60 static signature rules (31 high/24 medium/5 low) scan plugin sources for malicious install scripts, credential theft, obfuscation, persistence and network callbacks before 'dsh plugin add', plus 12 destructive-command patterns and workspace-boundary checks against accidental deletion. | 安装安全闸门与数据保护闸:60 条静态签名规则(31 高危/24 中危/5 低危)扫描插件恶意行为、持久化与外传,12 种危险命令模式与边界检查拦截误删。
DSH microsandbox bundle: Linux microVM shell executor and ctx.sandbox provider (msb run) with workdir/env preservation, fail-closed probing, and shell-free argv construction
Codex-style Guardian reviews, audit, and manual override for DeepSeek Harness
Remote access login gate for the DeepSeek Harness Web UI: a password wall in front of non-loopback access, session cookies, an injected browser-side gate, and a Remote Access settings section. The /api RPC surface is authenticated by dsh itself.
Turn DeepSeek Harness sessions into redacted, README-ready animated demos — as a local CLI or DSH plugin.
Content-addressed proof for body-free DSH tool-result custody stages
DSH profile plugin: unconfined (GPU-capable) session sandbox plus per-operation user approval for writes outside the workspace or to protected files.
Access-password gateway plugin for DeepSeek Harness Web: every request on the gateway port requires authentication (no unauthenticated surface besides the login/setup page), first visit sets the password, and it can be rotated from the dedicated Security settings page.
Read-side SIEM forwarder for DeepSeek Harness: normalises session activity to OCSF and ships it
Share a selected range of chat messages as Markdown, HTML, or plain text (Web /share command, Header button, sidebar ... menu, redaction, auto-save)
Stent MXC sandbox sidecar for the DeepSeek Harness
Plugin insight center — one answer to '哪些值得装': requirement matching (plugin_guide), environment recipes (recipe), health scoring (plugin_rank), static security scanning (plugin_audit), and a single install verdict (plugin_verdict)
A DeepSeek Harness (DSH) plugin that adds an AI-adjudicated 'smart' approval mode: safe operations are auto-allowed via an isolated side-channel LLM judge, everything else falls back to the normal human approval popup.
DSH agent-fleet hygiene audit: credential-file permissions, embedded credentials in git remotes (masked), provider token-prefix literals. Read-only, zero-dependency, deterministic.
Solve the AI goldfish brain: a personal memory layer for DeepSeek Harness — preferences, project conventions, workflows, and error lessons, stored locally as auditable, evidence-backed Markdown/JSON. 解决 AI 金鱼脑:偏好、项目约定、工作方式与纠错教训的本地可审计记忆层。
Evidence-backed application-security assurance for DeepSeek Harness
Content-addressed proof that recorded DSH authorization decisions and effects agree
安全加固:威胁建模(STRIDE)+ 三层边界系统,任何接收用户输入/鉴权/存储敏感数据/集成外部服务的功能都按安全第一开发。受 addyosmani/agent-skills(88k★ MIT)启发。
AI Agent runtime authorization & evidence verification — tool-call GuardrailProvider, CCS 7-dimension verification standard, MCP/DSH security scanner, SSRF/command-injection/credential-exfil blocking with Ed25519 signed receipts.
Look up what a DeepSeek Harness plugin can actually do, from inside dsh.
Offline content-addressed proof for supplied license-obligation delivery closure
修复 sandbox_permissions 和 justification 字段在 pwsh/bash/fs/dsh-sandbox 中的 no-op 升级报错。当会话已是 danger-full-access 模式时,模型携带空 justification 或重申同一模式被拒的问题。