The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill represents an indirect prompt injection surface because it ingests untrusted data from external Jira issue fields (summaries, descriptions, and comments) while maintaining the capability to perform write operations on the same platform.
Ingestion points: External data enters the agent context via the list and view commands in scripts/jira, which fetch ticket details from the Jira REST API.
Boundary markers: The skill does not implement explicit delimiters or specific "ignore embedded instructions" warnings for the data retrieved from Jira before it is presented to the agent.
Capability inventory: The CLI script has network access via the requests library and supports state-changing actions such as create, comment, and transition through POST and PUT requests to the Jira API.
Sanitization: The _extract_text method in scripts/jira converts Atlassian Document Format (ADF) JSON structures into plain text, which provides a layer of sanitization for the displayed content.

jira