The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill retrieves and processes Jira issue data, which can contain instructions from untrusted sources.
Ingestion points: Content is ingested via tools such as JIRA_GET_ISSUE, JIRA_SEARCH_FOR_ISSUES_USING_JQL_POST, and JIRA_LIST_ISSUE_COMMENTS in SKILL.md.
Boundary markers: The skill advises manual verification of targets and payloads but does not employ structured delimiters or safety prompts for the ingested content.
Capability inventory: The skill possesses extensive write permissions, including JIRA_CREATE_ISSUE, JIRA_EDIT_ISSUE, JIRA_ADD_COMMENT, and JIRA_ADD_USERS_TO_PROJECT_ROLE.
Sanitization: No specific filtering or sanitization of external Jira data is mentioned before processing.
[EXTERNAL_DOWNLOADS]: The skill configuration requires connecting to an external tool provider.
The setup instructions specify adding https://rube.app/mcp as a server, which relies on an external domain to provide tool definitions and execution capabilities.

jira-automation