The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requires the user to execute modal deploy execution/modal_webhook.py. This command pushes local code to a cloud environment for execution.
[EXTERNAL_DOWNLOADS]: The skill references endpoints on the modal.run domain. This is a well-known cloud computing service used for the skill's primary functionality.
[CREDENTIALS_UNSAFE]: The documentation specifies that MODAL_TOKEN_ID and MODAL_TOKEN_SECRET are required from the Modal dashboard. These are necessary for authenticated communication with the platform and are not hardcoded within the skill.
[REMOTE_CODE_EXECUTION]: The skill is designed to facilitate the execution of logic on remote Modal infrastructure via webhook triggers.
[PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection as it processes data from directive files and webhook inputs.
Ingestion points: directives/*.md files and the slug query parameter in the execution endpoint.
Boundary markers: No explicit delimiters or instructions to ignore embedded commands are present in the provided template.
Capability inventory: Includes code deployment (modal deploy) and tool usage (send_email, read_sheet, update_sheet).
Sanitization: No sanitization or validation logic is specified for the inputs or directive content.

add-webhook