jhckevin/dsh-auto-review: Automatic Approval and Review Plugin for DeepSeek Harness
`dsh-auto-review` is a security enhancement plugin for DeepSeek Harness, designed to address the conflict between automation efficiency and execution safety. It uses an independent review model to perform a secondary assessment of high-risk operations, while regular operations continue through the original workflow. The plugin features a Codex-style review interface, native sandbox isolation, a risk-tiering model, and circuit breaker mechanisms. Installation requires configuring the review model and initializing native components. Note that reviews incur additional costs, and isolation relies on the native bridge; do not rely solely on partial execution.
Read MoreDeepSeek Harness Automatic Approval and Review Plugin (dsh-auto-review)
`dsh-auto-review` is an automatic review plugin (v0.4.2) for DeepSeek Harness, designed to address pain points in tool invocation and permission management. Through an independent review layer, it distinguishes server-side failures (such as timeouts and errors) from dangerous operations, defaulting to manual approval rather than rejection. The plugin provides a 90-second total deadline, a complete review history, and fine-grained rule management (such as filtering by session, tool, and parameter). After installation, users can select “Approve for Me” in a session, leveraging Codex-style mechanisms to keep operations controllable and secure.
Read Moredsh-auto-review: Adding Guardian Automatic Approval to DeepSeek Harness
`dsh-auto-review` is a plugin for DeepSeek Harness (DSH) designed to automate security permission review. It introduces a Guardian model that automatically approves Shell, network, and MCP operations through the “auto-review” preset while allowing access to read-only workspaces. The plugin supports UI configuration and environment variable setup, and features audit logging, UI optimization, and manual override. When using it, note the “Fail Closed” policy and permission restrictions, and ensure the Node.js environment is installed and the corresponding installation and build commands are executed.
Read Moredsh-auto-review: DSH's default-enabled security review plugin
dsh-auto-review is the native security audit plugin for DSH (DeepSeek Harness), designed to decouple "permission control" from "content auditing" to prevent security reviews from being bypassed due to full access permissions. The plugin possesses two core capabilities: first, it performs local rule interception before executing high-risk operations such as write, edit, and Bash, adopting a default "fail-closed" strategy that denies execution until explicitly reviewed; and second, during change delivery, it conducts a review via an independent read-only sub-agent. If high-risk issues are detected, it defaults to requiring a fix; it will not falsely report as "clean" if the audit is interrupted or the results are unparseable. The plugin offers a session-level Auto Mode that is enabled by default and persists across restarts, which can be temporarily disabled in the current session via a command. It supports Web and cc-tui profiles, requires installation within the DSH environment, and requires Node.js 18.
Read Moredsh-auto-review: Automatic Approval Decision Plugin for DeepSeek Harness
`dsh-auto-review` is an MIT-licensed plugin for DeepSeek Harness, specifically designed to address the sandbox upgrade approval issue for actions exceeding `workspace-write` permissions when the `approval policy` is set to `'auto'`. It acts as an answerer for `approval/request` and utilizes a dual mechanism of "deterministic filtering + LLM review" to achieve automatic decision-making without human intervention. Core capabilities include: first, performing deterministic filtering to immediately reject or allow clear commands via `hardline` and regex rules; gray-area requests then undergo a clean-context LLM security review, making judgments based solely on the tool name, reason, and command to avoid carrying full session context. The plugin supports...
Read More