The Agent Skills Directory

[SAFE]: The skill contains only natural language instructions and structural guidelines for manuscript evaluation. No executable code, network exfiltration, or obfuscation techniques were found. The use of relative paths for internal resources is consistent with standard platform documentation practices.
[INDIRECT_PROMPT_INJECTION]: The skill establishes an ingestion surface by instructing the agent to read internal resource files (../../resources/source-basis.md) and search external websites for live author guidelines. While this exposes the agent to untrusted data from external sites, the risk is mitigated by instructions that prioritize official journal documentation over the skill's own heuristics. No dangerous capabilities such as shell execution, file-writing, or network exfiltration tools are utilized in the provided content.

the-innovation