tanner-code-review
Pass
Audited by Gen Agent Trust Hub on Jul 9, 2026
Risk Level: SAFEPROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill is designed to process untrusted data from
git diffoutputs and rendered web content, which presents an attack surface for indirect prompt injection where malicious instructions could be embedded in the reviewed code or site. - Ingestion points: Git diff outputs and web page content rendered in a headless browser (SKILL.md).
- Boundary markers: The instructions do not define delimiters or specific warnings for the agent to disregard instructions within the data being reviewed.
- Capability inventory: The skill uses shell commands (
git diff,grep) and browser automation tools (SKILL.md). - Sanitization: There are no mentions of sanitizing or filtering the external content before processing.
- [COMMAND_EXECUTION]: The skill utilizes standard command-line tools, specifically
git diffandgrep, to extract and map changed code within the repository. - [EXTERNAL_DOWNLOADS]: The skill directs the agent to access external production URLs and local server routes via a headless browser to perform visual and accessibility audits.
Audit Metadata