jaccen/Awesome-Gaussian-Skills

C3

1000+ arXiv-verified 3DGS methods · 15 AI agent skills · MCP server (34 tools) · open ebook — the AI-native entry to 3D Gaussian Splatting research

★ 100+ · jaccen/Awesome-Gaussian-Skills source on GitHub · this plugin in the registry

jaccen/Awesome-Gaussian-Skills is a DeepSeek Harness plugin rated C3 — powerful capability combined with sensitive behavior. It patches the dsh runtime, executes system commands, uses eval / new Function, decodes base64 payloads.

Installable plugin — declares a dsh.bundle manifest

What it can do

CapabilityFlagEvidence
patches the dsh runtimeruntime_patch./cordis.patch.yml
executes system commandsexec×9 in authored code, e.g. scripts/build-dsh-bundle.mjs:21, scripts/build-dsh-bundle.mjs:21
uses eval / new Functioneval×1 in authored code, e.g. studio/bridge/src/pipeline/pipeline-config.ts:173
decodes base64 payloadsbase64_decode×4 in authored code, e.g. mcp-server/src/sculpt.ts:859, mcp-server/src/tools.ts:439
starts a network servernet_server×1 in authored code, e.g. mcp-server/src/renderer-bridge.ts:64
reads credential-class env varstoken_envASR_ALIYUN_APPKEY, ASR_ALIYUN_KEY, ASR_TENCENT_SECRET_ID, ASR_TENCENT_SECRET_KEY, COSYVOICE_API_KEY, KLING_API_KEY

Tool registrations

0 confirmed registration point(s); analysis is incomplete.

1 source file(s) omitted by size/count/encoding limits; tool count is incomplete

Outbound domains

api.deepseek.com api.seedance.ai unpkg.com nls-gateway.cn-shanghai.aliyuncs.com api.openai.com api.kuaishou.com dashscope.aliyuncs.com api.moonshot.cn open.bigmodel.cn

Environment variables it reads

PIPELINE_OUTPUT_DIR LLM_API_KEY TOONFLOW_URL FFMPEG_PATH LLM_BASE_URL INCLUDE_EXPERIMENTAL ASR_WHISPER_MODEL LLM_MODEL MPT_API_URL SCENE_DIRS

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.