The Agent Skills Directory

[COMMAND_EXECUTION]: The skill provides explicit instructions for the agent to execute a shell script named independent_review.sh from the repository's local skills/ directory. This script is used as an optional escalation for high-risk reviews and accepts a file path as a command-line argument.
[INDIRECT_PROMPT_INJECTION]: The skill analyzes potentially untrusted data found in the website's source code, configuration files, and documentation during the audit process.
Ingestion points: Audits project files including source code (src/pages/**), configuration files (src/config.ts, _headers, _redirects), and markdown documentation.
Boundary markers: The instructions do not specify the use of delimiters or instructions to ignore embedded prompts within the analyzed files.
Capability inventory: The agent is instructed to perform code reviews using internal tools and execute local shell scripts based on findings.
Sanitization: No sanitization or filtering logic is specified for the content extracted from the files before it is processed by the agent.

website-review