The Agent Skills Directory

[COMMAND_EXECUTION]: The skill primarily functions by executing the af jira command-line tool. This includes commands that perform read and write operations on a Jira instance based on user or external input.- [DATA_EXFILTRATION]: The af jira attach command allows for the uploading of local files to a remote Jira server. While this is a standard feature for Jira management, it presents an exfiltration vector if an agent is manipulated into uploading sensitive system files (e.g., .env files, SSH keys, or configuration data) to a Jira issue.- [PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection because it ingests and processes content from an external source (Jira) that could contain malicious instructions.
Ingestion points: The skill retrieves untrusted data from Jira through operations like af jira get, af jira list, af jira search, and af jira comment. This data includes issue summaries, descriptions, and comments.
Boundary markers: There are no specified delimiters or system instructions defined within the skill to distinguish between administrative instructions and the data retrieved from Jira.
Capability inventory: The skill possesses significant capabilities, including creating, updating, and deleting Jira resources, as well as the ability to transmit local files to a remote network endpoint.
Sanitization: The skill description does not include any mechanisms for sanitizing or validating the content retrieved from Jira before it is interpreted by the agent.

jira