spec-to-backlog
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes potentially untrusted data from Confluence pages to generate content for Jira issues. The following analysis of the attack surface was performed:
- Ingestion points: Data enters the skill context through the
getConfluenceContenttool (Step 1, SKILL.md). - Boundary markers: The skill does not explicitly utilize delimiters or "ignore instructions" warnings for the ingested spec content.
- Capability inventory: The skill possesses write capabilities through the
createJiraIssuetool (Steps 5 and 6, SKILL.md). - Sanitization: There is no evidence of specific input sanitization before interpolation into issue summaries or descriptions.
- Risk Mitigation: The skill implements a robust human-in-the-loop mechanism in Step 4 ("Wait for user confirmation before proceeding"). This ensures the user reviews all proposed actions, effectively mitigating the risks associated with processing external content.
Audit Metadata