sentry-javascript-bugs
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 code from users or branch diffs to perform bug pattern analysis. This creates an attack surface where malicious instructions could be embedded in source code, though this is a standard characteristic for analysis tools. \n- Ingestion points: Untrusted source code and branch diffs provided for review (SKILL.md). \n- Boundary markers: None identified; instructions do not specify delimiters for data isolation. \n- Capability inventory: Access to file reading (Read, Glob, Grep) and shell utilities (Bash) (SKILL.md frontmatter). \n- Sanitization: Content is analyzed as source code; no filtering or sanitization of input text is performed.
Audit Metadata