sentry-backend-bugs
Pass
Audited by Gen Agent Trust Hub on Sep 17, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted code inputs, such as user-provided diffs and PR content, which is a necessary part of its code-review functionality but inherently introduces an indirect prompt injection surface.\n
- Ingestion points: untrusted code is ingested via user input or branch diffs as described in
SKILL.md.\n - Boundary markers: None present in the skill instructions to distinguish between code and potential embedded instructions.\n
- Capability inventory: The agent has access to
Read,Grep,Glob, andBashtools, as defined in theallowed-toolssection ofSKILL.md.\n - Sanitization: No input sanitization or filtering is defined within the skill's logic.\n- [SAFE]: No malicious patterns, obfuscation, or unauthorized access to sensitive information were found. The skill follows best practices for a documentation-heavy review tool, and its external references point to official Sentry repositories, which are trusted vendor resources.
Audit Metadata