The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from the local repository during its auditing phase.
Ingestion points: The agent is instructed to read monitoring configurations (Prometheus, Grafana, Datadog), existing alert files (alerts.yaml), and documentation (docs/, runbooks/) to understand the current state.
Boundary markers: There are no explicit instructions or delimiters defined to prevent the agent from obeying instructions that might be embedded within these audited files.
Capability inventory: The skill has access to sensitive tools including Bash, Write, and Edit, which could be used to modify the system or repository if the agent is misled by injected content.
Sanitization: No specific sanitization or validation steps are mentioned for the content read from the repository.

vigil-alert