The Agent Skills Directory

PROMPT_INJECTION (LOW): Indirect prompt injection surface detected. The skill processes ticket data from an external source (JIRA) which may be controlled by untrusted users.
Ingestion points: scripts/get_ticket.ts fetches and extracts text from the fields.description and fields.summary fields; scripts/list_tickets.ts fetches and displays fields.summary.
Boundary markers: Absent. The scripts output the raw text of the tickets directly to the console/context without using delimiters or system-level instructions to ignore potential commands embedded in the data.
Capability inventory: The skill possesses write capabilities including adding comments (scripts/add_comment.ts) and transitioning ticket statuses (scripts/update_ticket.ts).
Sanitization: Absent. The extractText function in get_ticket.ts performs structural extraction of the Atlassian Document Format (ADF) but does not sanitize the string content for malicious instructions.

jira