Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Benign
high confidencePurpose & Capability
Name/description (二次元同人、摄影插画、社区指引) match the SKILL.md which lists filtering options, return fields, example prompts, update cadence and user-facing notes. There are no unrelated resource requests (no cloud credentials, no system binaries).
Instruction Scope
SKILL.md contains only UI/response guidance (filter items, return fields, example queries, reminders). It does not instruct the agent to read local files, access environment variables, call external endpoints, or exfiltrate data.
Install Mechanism
No install specification and no code files — instruction-only skills present minimal surface area and nothing is written to disk.
Credentials
No environment variables, credentials, or config paths are requested; requested access is proportional to the described purpose.
Persistence & Privilege
always is false and model invocation is allowed (default). The skill does not request persistent privileges or system-level changes.
Assessment
This skill appears internally consistent and low-risk: it only describes how to present LOFTER-like features and asks for no secrets or installs. However, the skill’s source and homepage are unknown — if you plan to rely on it, test it with non-sensitive queries first, do not provide any credentials, and be aware that accuracy of recommendations/copyright info depends on the agent’s training and external data (the skill does not declare any verified data sources).Like a lobster shell, security has layers — review code before you run it.
latestvk97a73a5zfaw0zjg2dq1396sf583djeb
License
MIT-0
Free to use, modify, and redistribute. No attribution required.