marketing-brief
Pass
Audited by Gen Agent Trust Hub on Sep 28, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external sources such as GitHub PR descriptions, comments, and git commit history, creating a potential surface for indirect prompt injection attacks.
- Ingestion points: External data is ingested via
gh pr view,gh pr diff, andgit logcommands during the 'Discovery' phase inSKILL.md. - Boundary markers: There are no explicit instructions or delimiters defined to separate user-provided content from instructions, nor warnings to the agent to ignore instructions embedded in the analyzed code or PR data.
- Capability inventory: The skill utilizes shell commands (
gh,git), filesystem write operations to save the brief, and external tool usage (WebSearch). - Sanitization: The skill lacks validation or sanitization mechanisms for the text extracted from the codebase or PRs before processing it.
Audit Metadata