The Agent Skills Directory

[COMMAND_EXECUTION] (HIGH): The file scripts/create-guardrails-config.md uses the ! prefix to trigger shell command execution on the host system. It executes utilities such as grep, find, and date to dynamically populate configuration fields.
[CREDENTIALS_UNSAFE] (HIGH): The shell commands in scripts/create-guardrails-config.md are specifically designed to read from .env files, which are standard locations for storing sensitive credentials like API keys and database passwords. Accessing these files via shell grep operations is a dangerous pattern that can lead to credential leakage.
[PROMPT_INJECTION] (HIGH): The skill is vulnerable to Indirect Prompt Injection through the $ARGUMENTS parameter and the content of the files it reads. 1. Ingestion points: $ARGUMENTS variable and contents of .env files. 2. Boundary markers: No delimiters or instructions to ignore embedded commands are present in the processing logic. 3. Capability inventory: The skill possesses the capability to execute arbitrary shell commands via the ! syntax. 4. Sanitization: There is no evidence of sanitization or validation for the input data before it is processed by the shell or included in the output.

advanced-guardrails