The Agent Skills Directory

[SAFE]: The skill is designed for text transformation within an Obsidian vault environment. Analysis of SKILL.md and formulaFixer.ts shows no evidence of malicious activity.
[DATA_EXFILTRATION]: The skill does not perform any network operations (e.g., fetch, curl, or WebSocket connections). All file operations are restricted to the local vault via the Obsidian app.vault API.
[REMOTE_CODE_EXECUTION]: No external dependencies are imported beyond standard Obsidian types, and no dynamic code execution patterns like eval() or exec() are used.
[PROMPT_INJECTION]: The instructions do not contain attempts to override agent safety guidelines or manipulate the underlying LLM's system prompt.

notemdpro-formula-healer