frontend-review-security
Pass
Audited by Gen Agent Trust Hub on Oct 8, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill executes a local shell script,
scripts/audit-security.sh, which is intended to perform static analysis on a user-provided repository. This execution is a core part of the skill's auditing functionality and utilizes scripts provided by the skill author. - [INDIRECT_PROMPT_INJECTION]: The skill ingests and analyzes external source code from a user-specified repository, creating a potential surface for indirect prompt injection if the audited code contains instructions designed to manipulate the agent.
- Ingestion points: The agent reads source code from the target
<client-repo>and processes data fromraw/security.json. - Boundary markers: The instructions explicitly direct the agent not to execute
gh issue createcommands, ensuring that the final action remains under human control. - Capability inventory: The skill utilizes the
scripts/audit-security.shtool to perform its primary tasks. - Sanitization: The skill performs a raw review of the source code as required for security auditing, without specific sanitization of the input content.
Audit Metadata