The Agent Skills Directory

[Indirect Prompt Injection] (LOW): The skill is designed to retrieve and display full details of JIRA issues (summaries and descriptions), which are untrusted external inputs. If an issue contains malicious instructions, the agent might inadvertently execute them.
Ingestion points: Data enters the agent context through the jira-as issue get command mentioned in SKILL.md.
Boundary markers: There are no instructions in SKILL.md to wrap the retrieved JIRA content in delimiters or to warn the agent to ignore embedded instructions.
Capability inventory: The skill is granted Bash, Read, Glob, and Grep tool access in the YAML frontmatter of SKILL.md.
Sanitization: No sanitization or validation of the JIRA issue content is performed before it is presented to the agent.
[Persistence Mechanisms] (SAFE): The documentation in SKILL.md suggests that users add eval commands to their shell configuration files (.bashrc, .zshrc) to enable shell completion. While this is a common pattern for CLI tools, it is a manual user action and not an automated persistence attempt by the skill itself.
[Command Execution] (SAFE): The skill uses a custom CLI tool jira-as. This is expected behavior for a JIRA management skill and is restricted to the tools explicitly allowed in the metadata.

jira-issue-management