The Agent Skills Directory

[SAFE]: The skill provides a secure framework for using the ankitpokhrel/jira-cli tool. It mandates security practices such as avoiding token exposure and requiring user confirmation for data-modifying actions. Additionally, it follow best practices for secret management by instructing the user to use environment variables for credentials.
[EXTERNAL_DOWNLOADS]: The documentation points to the official JiraCLI installation guide on GitHub, which is a well-known and trusted source for this utility.
[PROMPT_INJECTION]: The skill has an attack surface for indirect prompt injection because it processes untrusted data from external Jira instances.
Ingestion points: Data enters the context through jira issue view and jira issue list in SKILL.md.
Boundary markers: The skill does not define specific delimiters for Jira content in the agent's context.
Capability inventory: The agent can perform various operations via the jira CLI, such as creating, transitioning, or deleting issues as seen in SKILL.md.
Sanitization: No explicit filtering or sanitization of the retrieved Jira data is performed.
Mitigation: The skill reduces the impact of potential injection by enforcing a "read-only first" workflow and requiring explicit user authorization for any write operations.

jira-cli