dsh-guardwall
Security for DeepSeek Harness in two layers: source-backed pre-install vetting plus fail-closed runtime guardrails and HMAC-chained audit logs. Zero runtime dependencies.
399 results
Security for DeepSeek Harness in two layers: source-backed pre-install vetting plus fail-closed runtime guardrails and HMAC-chained audit logs. Zero runtime dependencies.
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.
DSH AI 代码安全审查插件:secure_scan/secure_diff/secure_fix_verify/secure_report/secure_export/secure_baseline/secure_deps/secure_policy_show/secure_policy_set 九工具,40+ 确定性规则、密钥熵检测、git diff/staged 审查、SARIF 导出、基线接受与 SBOM-lite,零运行时依赖。
DSH 规则执行引擎 v3:容器解析 AGENTS.md + 理解器 + 匹配机 + 执行框架
Cybersecurity and reverse-engineering skill router for coding agents (pi / DSH / Codex CLI / Claude Code). Explicitly invoke when reverse-engineering, security analysis, CTF, or pentesting workflow is required.
Real-time attention alerts for DeepSeek Harness: a global banner plus flashing workspace entries when the agent is blocked (approval request / sandbox denial), with an audible chime and OS-level notifications for new blockers.
Export a minimal, secret-scrubbed, replayable problem bundle for DeepSeek Harness via the /repro command.
Scan agent skill packages for static security risks. Use for skill audit, security check, prompt-injection review, suspicious scripts, unsafe dependencies, zip package review, and pre-install skill review. Not for runtime monitoring.
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.
A safety harness plugin for DeepSeek Harness (DSH): protected-path enforcement, trash-based safe_delete with undo, last-known-good composition snapshots, pre-restart composition checks, and an audit journal.
Disk-usage audit for DeepSeek Harness (dsh) data directories: total size, per-directory breakdown, largest files, and oversized-file warnings (session logs can hit hundreds of MB). Zero runtime dependencies, read-only. CLI + agent-callable disk_audit tool.
A security-conscious registry and installer for external CLI tools exposed through DeepSeek Harness.
工具调用规范守卫:对 agent 工具调用参数做字符串匹配,命中危险行为拦截并注入原因(deny),或放行但注入警告(warn),附规则管理面板
Fail-closed DSH compatibility guard for redundant GPT/Codex sandbox escalation arguments
DeepSeek Harness plugin: auto-detect secrets pasted into the composer, store them in the official credentials seam, and send [secret:REF] placeholders to the model instead of the value.
Windows Git Bash fix for DeepSeek Harness (dsh): MSYS sandbox conflict + terminal inspector, plus the minimal-win collective-thinking preset (we/let's reasoning chain).
Native memory for DeepSeek Harness — auditable facts with evidence chains, powered by StateCore
Harness Plugin Store Optimized: a customized fork of dsh-plugin-hub — configurable plugin download directory, bundle-layer conflict detection, automatic pnpm lookup, and upstream bug fixes
Scan installed DeepSeek Harness plugins and grade stability risk (hook surface, startup work, preflight health, packaging, dependencies). 扫描已安装插件的稳定性风险(钩子面/启动任务/预检/打包/依赖)。
PRO-LONG-style programmatic memory for DeepSeek Harness: appends every session event to a per-workspace log.txt that the agent retrieves with grep/python. Built-in write probe, /prolong status command and permission-deny counter.
Independent Codex Guardian-style approval reviewer for DSH.
LLM pre-review for sandbox-escalation approvals: an independent-context LLM gate answers sandbox escalation requests before they reach the user, falling back to the user on any failure.
DeepSeek Harness (DSH) host-layer plugin: detects and auto-fixes Windows PowerShell command syntax mistakes — bash constructs, broken line continuations, pasted prompts, CRLF pollution — then executes the corrected command through the host shell seam under the normal sandbox/approval policy
WeCom (Enterprise WeChat) Smart Robot bridge for DeepSeek Harness — two-way chat over the aibot WebSocket gateway (bot_id + secret), no public endpoint needed.