jira-wiki-markup
Pass
Audited by Gen Agent Trust Hub on Sep 15, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADSNO_CODE
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill defines a 'Canonical repair-comment skeleton' in
reference.mdfor formatting automated reports. This template interpolates several variables (e.g.,{root_cause},{extra},{steps}) that are likely to contain untrusted data from external sources such as pull requests or commit messages.\n - Ingestion points: Data provided by host skills or external systems (branch names, PR descriptions, analysis summaries) are used to populate the report skeleton in
reference.md.\n - Boundary markers: The skeleton does not include specific delimiters or 'ignore' instructions to isolate interpolated values from the rest of the prompt.\n
- Capability inventory: The skill itself has no active tool permissions or executable scripts across its files.\n
- Sanitization: The skill provides no instructions for sanitizing or escaping the content of these variables before they are posted to Jira.\n- [EXTERNAL_DOWNLOADS]: The documentation in
SKILL.mdincludes an installation commandnpx skills add FuDesign2008/open-skillswhich references the vendor's repository for skill management.\n- [NO_CODE]: The skill is composed entirely of informational Markdown files and does not include any Python, Node.js, or shell scripts.
Audit Metadata