Development & Runtime

What it is — dsh-routing-suite provides runtime injector and router-standard presets.

Who it is for — You can install this if you need to use router-standard task-aware reasoning-mode router presets in DeepSeek Harness. You can install this if you need to use router-spec presets in DeepSeek Harness.

Watch out — It requires source installation from GitHub. Sandbox testing passed when registered in a new profile. Static checks found no risks.

The verdict — I would install it because it provides the injector and router presets for DeepSeek Harness agent management.

What it is — mirage provides a unified virtual filesystem for AI agents, allowing bash tools to operate across mounted resources.

Who it is for — It is suitable for AI agents that use bash tools and need to access multiple backends such as S3 and Slack at the same time. It is not suitable for agents that only run on local filesystems without remote mounted services.

Watch out — No obvious issues found. There is a static check prompt indicating it will execute shell commands. Sandbox testing was not performed.

The verdict — I would not install it because sandbox testing was not performed.

What it is — oh-dsh provides a DSH runtime bundle for Desktop, Web and TUI interfaces, including Workspace, PTY terminal, Git Review and plugin market.

Who it is for — If you use DeepSeek Harness for local development and need unified cross-end interfaces and shared plugin ecosystem, then oh-dsh is suitable. If you mainly work in a single interface environment without needing multi-end consistency, then you can install the TUI-only or Web-only distribution.

Watch out — Sandbox test passed after installation into a fresh profile and registration in harness. No obvious issues found. Source installation requires manual allowance of the build script.

The verdict — I would install it because it unifies the three interfaces and local tools integration; the premise is accepting source installation from GitHub.

What it is — pi2dsh is a Pi host ABI compatibility engine that mounts unmodified Pi npm packages as native DSH plugins on DeepSeek Harness.

Who it is for — Pi plugin developers who want to run their npm packages unmodified on DeepSeek Harness should install it. Developers who are not involved in Pi plugin development do not need to install it.

Watch out — It was tested successfully in a new profile and registered into the profile. A restart of DeepSeek Harness is required. Installation may require manually approving build scripts if pnpm blocks them.

The verdict — I would install it if I need Pi ecosystem plugins on DeepSeek Harness because it provides unmodified compatibility.

What it is — DSH plugin providing undo/redo/rollback for config changes and plugin code, with WebUI and offline CLI/GUI support.

Who it is for — If you frequently modify config and plugin code in DeepSeek Harness and want to undo those changes via WebUI or offline tools, consider this plugin. If your DeepSeek Harness usage does not involve such modifications, you may not need this plugin.

Watch out — Sandbox test passed: installed in a fresh profile and registered by harness. No obvious pitfalls found.

The verdict — I would install it because it provides offline tools as a fallback when DSH fails to boot.

What it is — DSH plugin that adds skill and MCP management panels to Web settings and provides unified dsh-panel CLI. Supports hot enable/disable, batch migration, drag-and-drop add, scoped operations, and CRUD for stdio/HTTP MCP servers.

Who it is for — If you are developing skills or configuring custom MCP servers in a DeepSeek Harness profile, this lets you handle everything directly in the settings page; if you only use built-in tools without needing extra panels or CLI subcommands, you do not need to install it.

Watch out — Installation from source requires manually enabling the build script. Sandbox testing passed with no other obvious issues found.

The verdict — Install if you need skill and MCP management integration; otherwise it is not worth it. The premise is that your profile involves more than just built-in tools.

What it is — Display DSH process network path graphs and support proxy configuration detection with snapshot-guarded repairs.

Who it is for — Users running DeepSeek Harness on Windows or WSL who need to diagnose network paths are suitable for installing this plugin. Users using macOS with stable network configurations without cross-model diagnosis needs are not suitable for installing this plugin.

Watch out — Sandbox test passed with no obvious issues found. Installation in a new profile succeeded and was registered by harness. Persistent modifications follow the snapshot-preview-confirmation-apply process.

The verdict — I would install it because it can generate Markdown network reports suitable for agents, which facilitates troubleshooting.

What it is — On Windows, this makes Minimal mode and persistent shell work in DeepSeek Harness by supplying the missing win32 process inspector so the persistent shell starts, using a busybox-based preset that survives workspace-write restricted token (MSYS shells die there), adding ConPTY console-list foreground resolution and GBK/UTF-16 file reads.

Who it is for — When using DeepSeek Harness in sandboxed mode on Windows and needing persistent shell survival, that scenario is suitable for installing it. If you run it on non-Windows platforms, you don't need to.

Watch out — The real cost is requiring Git for Windows. Sandbox testing passed: it was installed in a fresh profile and registered by harness. No other obvious issues found.

The verdict — If you need persistent shell in the sandbox on Windows with DeepSeek Harness, install it; otherwise, it's not worth it as the features are Windows-specific.

What it is — It runs multiple skills in parallel on the same task, generates a local HTML report, and lets you choose the winner to continue execution.

Who it is for — In academic polishing, SCI writing, PPTX generation or image prompt design tasks, having the agent compare multiple skill outputs in parallel can help you pick the winner. For tasks where the answer or tool path is already clear, people who can execute directly can skip it because direct execution will be faster.

Watch out — Sandbox testing passed; it was installed in a fresh profile and registered into it. Static checks indicate it will execute shell commands, but no other issues found. Installation from source requires manually allowing the build script.

The verdict — I would install it because it turns skill selection into an observable parallel execution process, but only when deliverables need file previews and QA.

What it is — The custom-first-control-prompt plugin injects deployment-configured system-prompt sections and reference exchanges into every conversation request via stream interception, with a web settings panel.

Who it is for — When you run long conversations in a specific model under DeepSeek Harness and need precise ordering of system prompt sections plus injected reference exchanges, this plugin fits you. It renders sections at the configured order position and prepends history as real messages into the request path.

If you do not care about system prompt ordering or reference injection, you do not need to install it.

Watch out — Sandbox testing confirmed successful installation and registration into the profile.

Configuration requires keeping section text non-volatile to preserve prefix-cache reuse.

The verdict — I would install it because it injects configuration content into every request path with zero log writes.

What it is — This plugin provides a management panel for the DeepSeek Harness MCP client, enabling inspection of connection status, registered tools, errors, and reconnect counts through the /mcp command and Settings tab.

Who it is for — When you need to add, edit, or remove MCP server configurations in DeepSeek Harness using the MCP client, you can install this plugin because it offers a Settings tab for those operations and supports tool trial calls. If you only connect a single static MCP server with infrequent changes, you can manage it directly in cordis.yml without installing this plugin.

Watch out — Sandbox testing passed when installed into a fresh profile and registered by Harness. No obvious issues were found.

The verdict — I would install it because it enables trial calls of MCP tools through the official execution pipeline with permission and approval guards, but only when profile modifications require explicit user approval.

What it is — Private bridge to run dsh runtime on Multica

Who it is for — If you need to run DeepSeek Harness tasks in the Multica environment, this plugin is suitable for you. If you are not running tasks in the Multica platform, it may not be relevant.

Watch out — Installation from source requires manually allowing the build script. Sandbox test passed in a new profile. No obvious pitfalls found.

The verdict — I would install it because it bridges the dsh runtime to Multica without modifying the Harness core and supports cooperative cancellation and durable session resume.

What it is — This is a private runtime bridge between Multica and DeepSeek Harness, exposing a versioned JSONL protocol over stdio.

Who it is for — Users integrating Multica runtime bridge in DeepSeek Harness profiles should use this plugin. If your workflow does not involve Multica's MCP configuration, you do not need this plugin because the runtime contract does not apply.

Watch out — Sandbox testing succeeded: the plugin installs successfully in a fresh profile and registers with harness. The static check indicates execution of shell commands. No other issues observed.

The verdict — I would install it because it enables Multica's task token forwarding into DSH shell execution, but only when using a Multica-specific profile.

What it is — Provides Huawei Cloud MCP service, skills and safety hooks for DeepSeek Harness.

Who it is for — When using DeepSeek Harness for coding agent tasks, users who need Huawei Cloud knowledge and safety hooks can use this plugin. Users who already have all necessary skills and hooks can choose other plugins.

Watch out — Sandbox test shows it can be installed but build scripts are blocked by pnpm: better-sqlite3 version 13.0.3 and huaweicloud-devkit version 1.1.0, leading to harness not completing registration. Users need to manually allow the build to successfully install. The installer automatically executes scripts and executes shell commands, but no other obvious pitfalls found.

The verdict — I would install it because it provides the necessary integration for DeepSeek Harness, but I would need to manually handle the build script permission first.

What it is — It is infrastructure providing a thin browser console and make-dsh-plugin skill for the DeepSeek Harness plugin ecosystem.

Who it is for — If you need to manage plugin installation states in your DeepSeek Harness profile and want the make-dsh-plugin skill for developing official plugins, this fits.

If you only use existing plugins without needing management or development tools, you don't need this because it focuses on ecosystem infrastructure rather than direct runtime features.

Watch out — It was tested to install successfully in a fresh profile and register with harness.

Note the specific #path: format required for git source installation on Windows to prevent command splitting.

No other obvious issues found.

The verdict — I wouldn't install it because it serves as infrastructure rather than a functional plugin; if your use case doesn't involve plugin development or state management, it's not worth it.

What it is — AgentDebugX plugin diagnoses DeepSeek Harness agent trajectories and generates structured reports.

Who it is for — When building multi-agent systems or computer-use agents, this plugin helps diagnose DeepSeek Harness trajectories. If you use other LLM frameworks, this plugin provides DSH specific debugging functionality.

Watch out — From source integration you need to manually allow the build script first. Static check found it executes shell commands. No obvious issues found.

The verdict — If agent development requires validating fixes, I would install this plugin because it supports reruns to test recovery proposals.

What it is — DSH Skills Manager manages local skills and inspects shared Agent skills in DeepSeek Harness.

Who it is for — When you need to manage skills in the Settings → Skills section of DeepSeek Harness web profile, this plugin is suitable. If you do not require importing a plugin directory containing SKILL.md via native file picker, this plugin is not necessary.

Watch out — Sandbox test passed: installed into a new profile and registered by Harness. No obvious pitfalls found.

The verdict — I would install it because it allows confirmation of name collisions, but only if using a supported DSH Web installation.

What it is — Integrates TapTap Maker into DeepSeek Harness, providing bundled MCP tools, Maker workflow and feature skills, as well as a packaged CLI for project initialization and diagnostics.

Who it is for — When performing TapTap Maker project initialization and diagnostics with DeepSeek Harness, the integration provides workflow and skills. Users who do not need TapTap Maker integration can continue with other configurations, as facts do not provide unique advantage data.

Watch out — It installs as an ordinary dependency and does not integrate profile layer. It executes shell commands. No obvious pits were found.

The verdict — It is not a profile plugin but provides packaged CLI if project initialization is needed.

What it is — Context Doctor audits token costs and detects duplicates/conflicts in DeepSeek Harness injected context: instruction chains, skill catalogs, tool schemas, and MCP tools.

Who it is for — When using DeepSeek Harness to manage multi-layer instruction chains and skill catalogs, you need to quantify token costs per item and identify name conflicts for the same skill; this plugin fits.

If you do not require quantification of injected context token costs or conflict detection, you need not install this plugin.

Watch out — Sandbox test passed: installed into a fresh profile and registered by harness. No obvious issues found.

The verdict — Install it because it supplies an executable context_audit tool and native panel that emit audit reports with executable trimming suggestions.

What it is — This is a zero-dependency read-only DeepSeek Harness plugin health check tool that scans plugin repositories and outputs compliance reports on manifest protocols, patch formats and build traps.

Who it is for — When you maintain multiple DeepSeek Harness plugin repositories in your organization and need regular checks on manifest protocols, patch formats and build traps, this tool fits your use case. If you only run a single plugin or your repository has already passed manual review, you do not need this tool.

Watch out — Sandbox testing passed: installed in a fresh profile and registered into harness. Static checks may mention executing shell commands but do not execute them. No other obvious issues found.

The verdict — I would install it because it provides automated compliance checks that reduce integration pitfalls in organizational plugin workflows, but only if installed via the profile bundle method.

What it is — This plugin converts DeepSeek Harness session, agent-loop, LLM, and tool lifecycle events into OpenTelemetry GenAI traces and metrics, exporting them over OTLP/HTTP to any compatible backend, with content capture off by default.

Who it is for — It suits users who want to observe agent loops and LLM/tool calls in DeepSeek Harness profiles and export traces to an OTLP backend. Users who do not need such observability do not need to install it.

Watch out — Sandbox testing confirmed successful installation and registration in a fresh profile. No obvious issues found. Static checks indicate shell command execution, but this is only a warning.

The verdict — I would install it only if I need to export DSH events to an OTLP backend; otherwise, not, since there is no need for observability.

What it is — Tool for backing up and restoring DeepSeek Harness user data with scheduled auto-backup and GitHub sync.

Who it is for — Users who need to back up and sync DeepSeek Harness user data across macOS, Linux, and Windows to a private GitHub repository. Users who do not have tar in their PATH are not suitable because the tool requires it.

Watch out — Sandbox testing passed and it can be registered in a new profile. Installation from source requires manually approving the build script. Backups contain plaintext credentials, so do not sync to untrusted locations.

The verdict — I would install it because it provides path-traversal rejection during restore and GitHub sync support, but only if using a private repository.