create-product-spec
Pass
Audited by Gen Agent Trust Hub on Nov 19, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection through untrusted GitHub issue data.
- Ingestion points: The skill reads input from
issue_context.jsonandissue_comments.txt(SKILL.md), which contain user-generated content from external GitHub issues and discussions. - Boundary markers: The instructions lack formal delimiters or explicit "ignore embedded instructions" warnings for the untrusted input, although they advise treating comments as "additional context" and "scratch input."
- Capability inventory: The skill is authorized to write product specification files in the
specs/directory and create a PR metadata JSON file. It is explicitly prohibited from using shell commands, performing git operations (commit, stage, push), or using the GitHub CLI. - Sanitization: No sanitization or escaping procedures are specified for the data ingested from the issue context or comments.
Audit Metadata