63 results
Agent-decided approvals for DeepSeek Harness: a workspace-write base permission mode where an independent approval subagent judges every sandbox escalation (risky operations are rejected), with a configurable approval model and a per-session audit trail in the conversation window's 审批 tab.
DeepSeek Harness bundle that routes real sandbox escalation prompts through a dedicated approval reviewer
DSH plugin: auto mode that routes permission-gated tool calls through an LLM review before approving, blocking, or asking for confirmation.
Pre-install supply-chain poison scanner for DeepSeek Harness plugins: AST analysis (NodeSecure JS-X-Ray) + deobfuscation decoder + regex heuristics to catch credential exfiltration, dynamic code execution, obfuscated imports, and install-time scripts.
DeepSeek Harness plugin (temporary, pre-upstream-fix): automatically injects a CPython tempfile shim (PYTHONPATH -> sitecustomize) into every confined shell command on the Windows sandbox, so python/pytest tempfile use works with zero extra tools, zero model-context overhead, and zero escalation.
Seatbelt (libsandbox) sandbox provider for the DeepSeek Harness sandbox seam on macOS: a cordis plugin that replaces @deepseek-ai/dsh-sandbox-local's seatbelt rung via a prebuilt seatbelt-run launcher — no dsh source changes, no PR required
Audit an agent harness against the harness-evaluation criteria, with machine-enforced evidence validation.
Forgetting engine for AI agents — memory TTL, decay, eviction, audit. The opposite of memory programming.
DSH Auto Guard: a Claude Code Auto Mode-like command approval mechanism that adds an LLM safety net on top of full access, with rules, caches, file tracker, and sensitive path gates.
dsh plugins ported from eco-agent: permission gate, audit chain, memory tree
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.
One-click health check & score for your DeepSeek Harness plugins. Scan → Score → Fix → Share. PerfScope for dsh.
OS-isolated capability capsules for third-party DeepSeek Harness plugins
Autonomous (auto) mode permission classifier for DeepSeek Harness: a Claude-Code-auto-mode-like classifier over tools/pre-execute and approval/request, a selectable 'auto' permission preset, LLM semantic judge, git checkpointing, agent discipline guidance, and a web control page in Settings → Plugins.
Pre-install supply-chain provenance checks for DeepSeek Harness plugins: verify the tarball you're about to install matches the source you read, before any code runs. Never executes the audited package.
One-click control that closes a session's persistent bash terminal (dual-face: /exit-bash command + session-header button) so the sandbox mode can be switched
Self-evolving memory for DeepSeek Harness (DSH): earned experiences, diary/fact semantic memory, concern tracking, and an append-only audit ledger.
DSH sandbox extension: configurable trusted writable roots outside the workspace, for both sandboxed CLI commands and the write/edit tools. 为 DSH 默认沙箱增加可配置的「沙箱授权目录」，允许工作区外的受信修改。
Password gate for dsh-web: opening the web port requires a username/password login; account credentials are stored salted+scrypt-hashed and AES-256-GCM encrypted in a local config file under $DSH_HOME. Host-only cordis plugin, no dsh source changes.
DSH profile plugin: unconfined (GPU-capable) session sandbox plus per-operation user approval for writes outside the workspace or to protected files.
【实验性】只读安全审计模式插件（自动只读、工具级白名单、强制交付选择）· EXPERIMENTAL plugin for the read-only security audit mode (auto read-only, tool allowlist gate, mandatory delivery choice)；日常使用推荐 dsh-presets 仓库的原生纯预设版。
Static and runtime security guard for the DeepSeek Harness: scans plugins and workspaces for malicious code, context injection and token waste (block/warn/clean), intercepts dangerous runtime tool calls and prompt steps, and exposes /scan, plugin_scan, the scan web panel and a user-managed allowlist. Fully static analysis — never executes scanned code.
DeepSeek Harness 多租户插件：认证反代网关（邮箱密码/LDAP/SSO）、角色权限（平台管理员/租户管理员/审计员/使用者）、token 配额与用量统计、审计日志
Natural-language driven dynamic assembler for DeepSeek Harness (dsh): discovers plugins at runtime (official-first, third-party optional), generates assembly plans, and loads them via Cordis — with built-in security audit gates for unofficial plugins.