alexdoandev/jev-laya-dsh

C3

User → Jev(Laya) → Agent — a local System-1 decision layer for DeepSeek Harness agents. Laya (open, Jev-compatible) routes prompts, gates risky tool calls and swaps models in milliseconds. 100% local,

★ 1+ · alexdoandev/jev-laya-dsh source on GitHub · this plugin in the registry

alexdoandev/jev-laya-dsh is a DeepSeek Harness plugin rated C3 — powerful capability combined with sensitive behavior. It runs code at install time, can rewrite the system prompt, can spawn subprocesses, gates tool execution.

Not installable — has plugin code but no manifest to install it by

What it can do

CapabilityFlagEvidence
runs code at install timeinstall_scriptpostinstall: node patches/apply-multilingual-patch.js
can rewrite the system promptprompt_surfacesubprocess, systemPrompt, tools/pre-execute
can spawn subprocessessubprocess_serviceinject: subprocess
gates tool executiontool_gatehook: tools/pre-execute
executes system commandsexec×5 in authored code, e.g. smoke-test.js:9, smoke-test.js:35
starts a network servernet_server×1 in authored code, e.g. onnx-server/laya-onnx-server.mjs:148
ships no manifestno_manifestuses 3 services, 0 tool regs, no manifest

Tool registrations

0 static tool registration point(s).

Services it injects

subprocess systemPrompt tools

Hooks it attaches

agent/pre-step agent/request api-session/added tools/pre-execute

Environment variables it reads

LAYA_PORT LAYA_IDLE_UNLOAD_SECS LAYA_MODEL_DIR LAYA_THREADS

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.