The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses bash -c to wrap curl commands. This is implemented as a workaround for environment variable handling in piped command sequences.
[EXTERNAL_DOWNLOADS]: The skill makes network requests to official JotForm domains (api.jotform.com, eu-api.jotform.com, hipaa-api.jotform.com) to retrieve and manage form data.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests untrusted data from external form submissions.
Ingestion points: Form submissions, questions, and properties retrieved via GET requests in SKILL.md.
Boundary markers: Absent. Data retrieved from the API is piped to jq and then directly into the agent's context without delimiters.
Capability inventory: The skill utilizes curl for network operations and bash -c for command execution, providing a surface for further actions if the agent is manipulated.
Sanitization: No explicit sanitization or filtering is performed on the data retrieved from JotForm before it is presented to the agent.

jotform