mcp-sentinel

Harness agent plugin that acts as a sentinel between the AI agent and MCP servers — polling long-running tasks so token-costly status loops never enter the LLM inference path

PluginAI Models

Verification

L2 · Structured

Security

Low

Health

Active

Trust

Silver

What it does

Harness agent plugin that acts as a sentinel between the AI agent and MCP servers — polling long-running tasks so token-costly status loops never enter the LLM inference path

Installation

dsh plugin --profile web add mcp-sentinel

Install method: npm · not yet tested in container (L3+)

Compatibility

DSH Version	Status
not stated	Declared — not tested

Requirements

• Node.js: not stated
• DSH: declared "not stated"
• External credentials: none detected

Security Report

Automated static scan, not manual review.

Risk: Low0 critical · 0 warning · 2 info

• Static heuristic scan: done (15 files)

• Dependency vulnerabilities: requires deep audit (L3+)

• Permission sandboxing: requires runtime testing (L4+)

Activity

Last commit 2026-08-15 · activity: active

6 mo

Source

GitHub: github.com/GCS-ZHN/mcp-sentinel