experience-lwc-rtl-validate
Pass
Audited by Gen Agent Trust Hub on Aug 19, 2026
Risk Level: SAFECOMMAND_EXECUTIONPROMPT_INJECTION
Full Analysis
- [PROMPT_INJECTION]: The skill processes user-supplied LWC source files (HTML, JS, CSS) as input. This creates an attack surface for indirect prompt injection where malicious instructions embedded in the code could attempt to influence the agent's behavior during the audit process.
- Ingestion points: The skill reads
.html,.js, and.cssfiles from the local filesystem (SKILL.md, Step 1). - Boundary markers: The skill instructs the agent to cite specific sections from
references/rtl-expert.mdin its report, which helps ground the output. - Capability inventory: The skill has the ability to execute a bundled shell script and write modifications to the component files (SKILL.md, Step 7).
- Sanitization: The
scripts/scan-rtl-css.shscript includes a function to strip CSS comments before scanning, reducing the injection surface for CSS files. - [COMMAND_EXECUTION]: The workflow involves executing a bundled script,
scripts/scan-rtl-css.sh, to perform automated CSS property scanning. This script executes an embedded Python program that uses regular expressions to identify and suggest logical property replacements.
Audit Metadata