The Agent Skills Directory

[DATA_EXFILTRATION]: The skill reads from ~/.config/.jira/.config.yml to extract the Jira project key. Accessing configuration files in the user's home directory is a sensitive file path access.
[COMMAND_EXECUTION]: Shell commands are used to interact with jira-cli and git. Examples include jira issue list and git log, which are executed to gather work activity metrics.
[EXTERNAL_DOWNLOADS]: The skill documentation refers to an external repository (github.com/ankitpokhrel/jira-cli) as a required dependency for the skill's functionality.
[PROMPT_INJECTION]: The skill possesses an indirect prompt injection attack surface:
Ingestion points: Untrusted data enters the agent context via Jira ticket summaries, descriptions, and git commit messages processed during the gathering phase.
Boundary markers: No explicit delimiters or instructions are used to separate untrusted data from the agent's core prompts when being analyzed by sub-agents in Phase 4.
Capability inventory: The skill uses cat for file reading and executes shell commands via the jira and git command-line tools.
Sanitization: There is no evidence of sanitization or escaping of the content retrieved from Jira or Git before it is interpolated into prompts for analysis.

jira-daily