relay-dsh-plugin-monitor-process
Run the following command in DeepSeek Harness:
dsh plugin install yangbobo2021/relay-dsh-plugin-monitor-process
Paste the following prompt into your AI chat to install this plugin:
Run dsh plugin install yangbobo2021/relay-dsh-plugin-monitor-process in the DeepSeek Harness terminal; the source is available at https://github.com/yangbobo2021/relay-dsh-plugin-monitor-process
About this plugin
In Relay monitoring pipelines, Agents frequently need read-only process status, yet exposing PIDs directly or embedding them into Bundles creates identity-leak and privilege-escalation risks.
This plugin closes that gap with a deliberately minimal, identity-safe capability. It ships no prebuilt Bundle Type; instead the Agent issues an opaque Process Handle bound to the current Session and Project. A PID is accepted only at the moment of handle issuance and is never embedded in a custom Bundle. The Agent then leverages the relay-monitor-author Skill to assemble a task-scoped, single-read Monitor Bundle under least privilege. The plugin is designed to run alongside Monitor Core and Author and remains compatible with the audited DSH 0.1.2-alpha.3 profile.
It targets Agent developers and SRE teams building Relay monitoring workflows on DSH 0.1.2-alpha.3 or later who depend on Monitor Core and Author and need process-level observability under strict identity isolation.
Use Cases
- Agent reads target process status without ever exposing a PID
- Assembling a task-scoped read-only Monitor Bundle bound to the current Session and Project
- Performing process observability alongside Monitor Core and Author under DSH 0.1.2-alpha.3
Best For
- Agent developers building Relay monitoring workflows
- SRE teams needing process-level observability with strict identity isolation
- Engineering teams running security audits on DSH 0.1.2-alpha.3 or later
Related Plugins
A network security red-team benchmark plugin for DeepSeek-V4.1/Flash featuring dual-layer kernel injection and zero-tool architecture for controlled model safety assessment.
A security red-team prompt plugin for DeepSeek-V4 that uses code-formal mapping and zero-trigger system prompts to enable unconditional output, intended for authorized testing and research only.
Adds an Auto approval mode on the official workspace-write sandbox, classifying semantic risks, asking when ambiguous, and denying destructive operations.
Undo/rollback system for DSH: config & plugin snapshots, one-click undo/redo/restore, message-level revert, secret masking, safe mode, and offline WebUI/GUI/CLI rescue tools for when DSH won't start.