The Agent Skills Directory

[COMMAND_EXECUTION]: The script scripts/jira-tool.sh constructs JSON payloads for several commands (e.g., cmd_comment, cmd_create, cmd_assign) by directly interpolating shell variables. Without consistent escaping for characters like double quotes, this can result in malformed JSON or the injection of additional fields into Jira API requests.
[COMMAND_EXECUTION]: The script invokes local browser openers such as open or xdg-open using URLs dynamically parsed from cloudflared output during the authentication flow.
[PROMPT_INJECTION]: The skill is susceptible to indirect prompt injection due to its handling of external data.
Ingestion points: Issue summaries, descriptions, and comments are retrieved from jira.cfdata.org via the get and search commands.
Boundary markers: No delimiters or safety instructions are applied to the retrieved content to prevent the agent from interpreting it as instructions.
Capability inventory: The skill has broad capabilities to modify the Jira environment, including creating, updating, and deleting issues.
Sanitization: There is no sanitization or validation of data retrieved from Jira before it is processed by the agent.

jira-tool