GCS-ZHN/mcp-sentinel
C3Harness 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
★ 5+ · GCS-ZHN/mcp-sentinel source on GitHub · this plugin in the registry
GCS-ZHN/mcp-sentinel is a DeepSeek Harness plugin rated C3 — powerful capability combined with sensitive behavior. It patches the dsh runtime, executes system commands, reads credential-class env vars.
What it can do
| Capability | Flag | Evidence |
|---|---|---|
| patches the dsh runtime | runtime_patch | ./cordis.patch.yml |
| executes system commands | exec | ×8 src, e.g. scripts/run-e2e.ts:34, scripts/run-e2e.ts:34 |
| reads credential-class env vars | token_env | DEEPSEEK_API_KEY |
Services it injects
agents tools
Outbound domains
api.deepseek.com modelcontextprotocol.io deepseek-harness.github.io
Environment variables it reads
DEEPSEEK_API_KEY USERPROFILE DEEPSEEK_BASE_URL E2E_JUDGE_MODEL CODEX_BIN CODEX_CLI_PATH
How to read this
Levels measure capability surface and transparency, not maliciousness. A C3 plugin can be entirely legitimate — a desktop shell genuinely needs subprocesses. The point is that you can see this before installing. See the levels explained and how dsh plugins work.
Findings come from static analysis of shipped code; nothing is executed. Think a flag is wrong? Open an issue — every flag cites the file and line it came from.