jira-story-point
Pass
Audited by Gen Agent Trust Hub on Sep 5, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted data from Jira issue fields like descriptions and acceptance criteria, which provides a surface for indirect prompt injection attacks where malicious instructions could be embedded in tickets.\n- Ingestion points: Jira issue-retrieval and search tools are used to bring external ticket data into the agent's context as specified in SKILL.md.\n- Boundary markers: The instructions lack explicit delimiters or guardrails to differentiate between system instructions and data fetched from Jira.\n- Capability inventory: The skill uses tools to read, search, and update Jira tickets, allowing it to potentially act on malicious instructions in the data.\n- Sanitization: There are no documented steps for the agent to sanitize or validate the content retrieved from Jira issues before processing it.
Audit Metadata