The Agent Skills Directory

[SAFE]: The skill consists exclusively of static Markdown documents providing guidelines and pitfalls for kitchen renovation. No dangerous functionality was identified during analysis.
[PROMPT_INJECTION]: The instructions in SKILL.md are used to define the agent's persona and provide a structured framework for answering user questions about home improvement. No bypass markers, override commands, or attempts to extract system prompts were detected.
[DATA_EXPOSURE_AND_EXFILTRATION]: There are no hardcoded credentials, API keys, or sensitive file paths (such as .ssh or .env) within the documentation. The skill does not contain any tools for network communication or data transmission.
[REMOTE_CODE_EXECUTION]: The skill does not include script files (.js, .py, .sh), package manifest files, or instructions to download and execute code from remote sources. All content is local and non-executable.
[INDIRECT_PROMPT_INJECTION]: While the skill processes user queries regarding kitchen renovation, it lacks the technical capabilities (network access, file system writes, or subprocess execution) to be exploited via data ingestion. The risk is negligible as it only generates text responses based on local knowledge files.
[OBFUSCATION]: All files were scanned for Base64 encoding, zero-width characters, and homoglyphs. No hidden or obfuscated content was found. All URLs and instructions are in plain text and consistent with the stated purpose of the skill.

chufang-bikeng-zhinan