The Agent Skills Directory

[COMMAND_EXECUTION]: The skill utilizes the pnpm action CLI tool to perform data operations such as list-responses, response-insights, preview-form, and export-responses. These commands are used to interact with the local application environment and database.
[DATA_EXPOSURE]: The skill accesses the responses SQL table, which contains sensitive information such as submitterEmail and ip (Submitter IP). This access is necessary for the skill's primary function of form response management and auditing.
[PROMPT_INJECTION]: The skill exhibits an indirect prompt injection surface as it processes untrusted user-generated content from form responses.
Ingestion points: The data column in the responses table, which is read and processed by list-responses and response-insights.
Boundary markers: No explicit boundary markers or delimiters are defined in the instructions to separate untrusted response data from agent instructions.
Capability inventory: The agent can read database records, perform calculations on the data, and write to the local filesystem via export-responses.
Sanitization: There are no instructions provided for sanitizing or escaping the content retrieved from the form submissions before analysis.

form-responses