The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to analyze container security, which involves ingesting untrusted data such as configuration manifests, logs, and environment metadata.
Ingestion points: The agent is expected to use Read and Bash to examine container-related files and runtime environments.
Boundary markers: The instructions lack specific guidance or delimiters to prevent the agent from following malicious instructions potentially embedded in the audited data.
Capability inventory: The skill possesses high-privilege capabilities including Bash(cmd:*), Write, and Edit, which could be exploited if the agent is manipulated by data it audits.
Sanitization: There is no mention of input validation or sanitization for the data being processed.
[COMMAND_EXECUTION]: The skill requests unrestricted Bash access (cmd:*). While this is typical for terminal-based agents, the trigger keywords are extremely broad (e.g., "container"), increasing the likelihood of the skill activating in contexts where arbitrary command execution might not be intended by the user.

container-security-auditor