lutrodev/dsh-roleplay

C3

Roleplay plugin suite for DeepSeek Harness: character cards, lorebooks, personas, presets, state, and conversation tools.

★ 5+ · lutrodev/dsh-roleplay source on GitHub · this plugin in the registry

lutrodev/dsh-roleplay is a DeepSeek Harness plugin rated C3 — powerful capability combined with sensitive behavior. It can rewrite the system prompt, executes system commands, eval in build output only, decodes base64 payloads.

Installable plugin — declares a dsh.bundle manifest

What it can do

CapabilityFlagEvidence
can rewrite the system promptprompt_surfacesystemPrompt, system-prompt/assemble
executes system commandsexec×9 in authored code, e.g. scripts/dev.mjs:3, scripts/dev.mjs:3
eval in build output onlyeval_bundled×1 in build output only, e.g. packages/rp-remote/dist/client.js:93
decodes base64 payloadsbase64_decode×5 in authored code, e.g. plugins/rp-character-card/src/character-card.js:365, plugins/rp-character-card/src/client.js:415

Services it injects

agentPresets agents commands conversation dshHomePath llm loader locale remote rpAssetEditors rpFeatures rpRemote rpRuntime rpSessions sessionPersistence sessions settingsScope slots subagents systemPrompt tokenMeter tools typert uiConversation workspaceRegistry workspaces

Hooks it attaches

agent/disposed agent/pre-step agent/request agent/session-start llm/stream loader/entry-init loader/partial-dispose session/disposed session/event system-prompt/assemble

Environment variables it reads

npm_execpath DSH_TELEMETRY_MODE DSH_TOOLS_MODE

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.