The Agent Skills Directory

[PROMPT_INJECTION]: The skill is vulnerable to indirect prompt injection. Ingestion points: The skill processes untrusted code from *.dart, *.html, *.ts, and *.scss files as part of its auditing process. Boundary markers: There are no explicit delimiters or instructions provided to the agent to ignore instructions embedded within the processed code. Capability inventory: The skill has access to Bash, Agent, Read, Grep, and Glob tools, enabling filesystem modifications and nested agent calls. Sanitization: No sanitization or validation of the ingested code content is performed.

fixing-accessibility