The Agent Skills Directory

[PRIVILEGE_ESCALATION]: The skill suggests using chmod -R 777 on the var/cache and var/log directories in reference.md. This grants universal read, write, and execute permissions to all users on the system, which is a significant security risk compared to using more restrictive permissions or Access Control Lists (ACLs).
[COMMAND_EXECUTION]: The skill instructs the agent to execute various shell commands documented in reference.md, including composer install, various bin/console commands, and arbitrary PHP code via php -r. These commands can execute code from the vendor directory or modify project files.
[INDIRECT_PROMPT_INJECTION]: The skill possesses a vulnerability surface by reading project-level configuration files.
Ingestion points: The skill reads .env and config/*.yaml files as specified in reference.md.
Boundary markers: No delimiters or instructions are provided to the agent to ignore potentially malicious content within these files.
Capability inventory: The skill has the capability to execute shell commands and file system operations (chmod, bin/console, composer).
Sanitization: There is no evidence of sanitization or validation of the content read from the configuration files before it is processed by the agent.

bootstrap-check