The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructs the user to execute modal deploy to push orchestration logic to the Modal cloud platform.
[INDIRECT_PROMPT_INJECTION]: The skill establishes an infrastructure for event-driven execution that ingests data from external webhook triggers, creating a surface for indirect injection.
Ingestion points: External HTTP requests targeted at the generated Modal webhook endpoints as described in SKILL.md.
Boundary markers: The process lacks explicit delimiters or instructions for the agent to ignore potentially malicious commands embedded in the event data.
Capability inventory: The resulting webhooks are granted access to sensitive tools including send_email, read_sheet, and update_sheet as specified in the webhooks.json configuration.
Sanitization: No input validation or sanitization logic is defined for the incoming payloads before they are processed by the agent orchestration logic.

add-webhook