Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Interactive Infographic
v0.1.8强大的AI自动化绘图与复杂信息可视化工具（基于 ContextWeave）。不仅支持代码与系统架构的可视化，更广泛适用于复杂逻辑梳理、知识库转换、业务流程图、思维导图及长文本的结构化信息图生成。通过深度的语义分析与请求编排，一键将晦涩文本与复杂知识转化为清晰直观的图形表达。
⭐ 0· 180·0 current·0 all-time
byContextWeave@qhyw99
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The skill's name/description (convert text into diagrams via ContextWeave backend) aligns with its code (node scripts that read an input file and POST to a remote /run endpoint). Requiring node and an API key is expected. However, the repository includes a sample config (cwmcp_config.example.json) suggesting an api_url override, but the runtime code does not read that config and instead hardcodes https://pptx.chenxitech.site. That mismatch is unexpected and reduces transparency.
Instruction Scope
Runtime instructions and scripts restrict file reads to the provided input_file and validate that it is absolute and inside the current working directory before reading. The scripts follow the described workflow (write file → run scripts/generate_contextweave.cjs → POST). The SKILL.md explicitly discloses that data (and API key) will be routed to the built-in server. One minor inconsistency: SKILL.md references reading CONTEXTWEAVE_EDITOR_PROTOCOL and the code does so, but that env var is not documented in the registry metadata.
Install Mechanism
No install spec is present (instruction + scripts only). This minimizes surprising installation behavior (nothing is downloaded or installed automatically). The risk comes from executing local node scripts that will make network calls — not from an install/download step.
Credentials
Declared primary credential is CONTEXTWEAVE_MCP_API_KEY (appropriate), but the code also reads CONTEXTWEAVE_EDITOR_PROTOCOL (not declared in metadata). The CW client contains a hard-coded default API key value embedded in source; this means the skill will operate even without a user-provided key and could cause requests to be attributed to that shared/anonymous key. The declared set of env vars is incomplete relative to what the code actually uses.
Persistence & Privilege
The skill is not always-enabled and does not request elevated or persistent platform privileges. It does not modify other skills or system-wide settings. Scripts write files into the working directory only and the client enforces path checks to avoid traversal outside the cwd.
What to consider before installing
This skill will send the text you provide (and any CW code blocks in the input file) to a remote server hard-coded as https://pptx.chenxitech.site. The code also contains a built-in default API key, and it reads an additional env var (CONTEXTWEAVE_EDITOR_PROTOCOL) that isn't listed in the metadata. Before using: (1) don't supply sensitive or confidential data unless you trust the remote domain and operator; (2) review and, if desired, change the hard-coded base URL and remove the embedded default API key or replace it with a key you control; (3) prefer running the scripts in a sandboxed environment and test with non-sensitive inputs; (4) be aware that some helper scripts default to relative names (e.g., import/export) but the client enforces absolute paths — expect some CLI rough edges. If you want to trust this skill, ask the publisher for a clear statement of the backend operator, or modify the scripts to point to a backend you control.Like a lobster shell, security has layers — review code before you run it.
latestvk978k1ab30k9c3j0pr6xqxtn3583kvsd
License
MIT-0
Free to use, modify, and redistribute. No attribution required.
Runtime requirements
🧠 Clawdis
Binsnode
Primary envCONTEXTWEAVE_MCP_API_KEY