The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes shell commands using osascript to run AppleScript and uses the tinyfish CLI for data research. It dynamically generates temporary scripts in /tmp to automate interactions with Microsoft Excel.
[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection (Category 8) as it processes untrusted data from two sources: active Excel workbooks and content fetched from the web.
Ingestion points: Data read from the active sheet in Step 1 and web content retrieved in Step 3.
Boundary markers: There are no delimiters or instructions to treat data from the spreadsheet or search results as untrusted content.
Capability inventory: The skill can write files, execute AppleScript, run Python code with subprocess access, and perform network lookups.
Sanitization: While the skill escapes characters to prevent script syntax errors, it does not sanitize retrieved data to prevent the execution of malicious instructions embedded in cells or web pages.

enrich-excel