The Agent Skills Directory

[DATA_EXFILTRATION]: The skill is designed to interact with and manage highly sensitive configuration files and secret storage.
Evidence: The skill specifically targets sensitive paths such as .env, .env.local, and the Symfony secrets vault located in config/secrets/ (reference.md).
Evidence: The documentation includes instructions for using the php bin/console secrets:list --reveal command, which intentionally exposes secret values to the current process context for debugging and verification purposes (reference.md).
[INDIRECT_PROMPT_INJECTION]: The skill possesses a surface for indirect prompt injection by consuming and processing external configuration files.
Ingestion points: The agent is instructed to read and modify .env files and YAML configuration files like services.yaml and those in config/packages/ (SKILL.md, reference.md).
Boundary markers: The skill does not provide specific instructions on using delimiters or boundary markers to differentiate between legitimate configuration data and potential malicious instructions embedded in these files.
Capability inventory: The skill has access to powerful operations including filesystem edits (Edit, Write) and shell command execution (Bash), which could be misused if the agent follows instructions found within untrusted configuration data (SKILL.md).
Sanitization: There are no documented validation or sanitization steps for the content of configuration files before they are read or processed by the agent.

config-env-parameters