rhdh-jira-create
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructions direct the agent to execute shell commands using the Atlassian CLI (acli) for creating work items, adding comments, and linking issues. These commands use parameters derived from user interviews and parent issue metadata.
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from user conversations and support cases to generate Jira issue content, creating an attack surface. 1. Ingestion points: User-provided text and support case data processed in workflows/create-issue.md and references/support-intake.md. 2. Boundary markers: Absent; there are no explicit delimiters or 'ignore' instructions specified for the ingested data. 3. Capability inventory: Shell command execution via acli and file system writing via mktemp. 4. Sanitization: The skill includes instructions to remove customer names from public fields and uses the external /rhdh-jira-api skill for ADF format conversion.
- [DATA_EXPOSURE]: The skill utilizes temporary filesystem paths like $TMPDIR or %TEMP% to store drafted issue content for user review. While part of the standard workflow, writing potentially sensitive business logic or issue details to shared temporary directories can pose a minor exposure risk depending on the host's configuration.
Audit Metadata