The Agent Skills Directory

[PROMPT_INJECTION]: The skill is susceptible to Indirect Prompt Injection because it ingest and processes untrusted data from external sources.
Ingestion points: The skill reads full email threads via gog gmail read and retrieves data from Google Sheets using gog sheets get.
Boundary markers: There are no delimiters or instructions provided to the agent to ignore or isolate potentially malicious instructions embedded within the emails or sheet data.
Capability inventory: The skill has the ability to read and write local files in the /Users/steipete/Projects/conferences directory, and it can update Google Sheets via gog sheets update.
Sanitization: There is no evidence of sanitization, validation, or filtering of the content retrieved from Gmail or Sheets before it is used to make decisions or update files.
[COMMAND_EXECUTION]: The skill makes extensive use of the gog command-line utility to interact with Google services. It also manages files within a local project directory (/Users/steipete/Projects/conferences). These capabilities are necessary for the skill's functionality but provide a broad attack surface if the agent's logic is subverted through a prompt injection attack.

speaking