openspace-llm-xiaowei
Security checks across malware telemetry and agentic risk
Overview
This is mostly a coherent MiniMax LLM wrapper, but its documentation exposes a real-looking API key and under-declares the credential requirement.
Use your own MiniMax API key, do not copy the key shown in the README, and verify the OpenSpace package and proxy settings before installing. Avoid sending confidential text unless you trust the configured provider and account.
VirusTotal
VirusTotal findings are pending for this skill version.
Risk analysis
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
Users may unknowingly copy or rely on a shared/exposed key, which can create account, billing, quota, privacy, and revocation risks.
The README contains a full, live-looking MiniMax API key instead of a placeholder, despite the registry metadata declaring no primary credential or required environment variables.
OPENSPACE_API_KEY=sk-cp-MaMBLuaFp3MYYpki0XWxiT7iTrQ9Qz-uTad...
Remove and rotate the exposed key, replace it with a placeholder, and declare OPENSPACE_API_KEY as a required credential in the skill metadata.
A changed or compromised dependency could affect the behavior of the skill when installed.
The skill depends on an external PyPI package without a pinned version or lockfile; this is expected for the OpenSpace integration but leaves package provenance/version to the user.
pip install openspace
Install from a trusted source, consider pinning a known-good version, and review the OpenSpace package before use.
Any text submitted to the skill may be processed by the external MiniMax/OpenSpace provider and possibly through the configured proxy.
User prompts and analyzed text are sent to the configured external LLM provider; this is the core purpose of the skill and tool execution is disabled.
DEFAULT_API_BASE = os.environ.get("OPENSPACE_API_BASE", "https://api.minimax.chat/v1") ... result = await client.complete(messages=messages, execute_tools=False)Do not submit secrets or sensitive private data unless you trust the provider, proxy, and account configuration.