The Agent Skills Directory

[COMMAND_EXECUTION]: The skill instructions direct the agent to execute glab mr create using a --title argument populated with data sourced from external ClickUp tasks or GitLab issues. Because the instruction suggests wrapping the title in double quotes (--title "<title>"), there is a risk of shell command injection if an attacker-controlled task or issue title contains shell metacharacters such as backticks or interpolation syntax (e.g., $(...)).
[INDIRECT_PROMPT_INJECTION]: The skill processes untrusted data from external platforms which could be used to influence the agent's actions or downstream processes.
Ingestion points: Reads ClickUp task names via clickup task and GitLab issue titles via glab issue view (SKILL.md).
Boundary markers: None present to isolate the external task/issue content from the rest of the command context.
Capability inventory: The skill has access to the Bash tool and performs repository operations including git push and glab mr create (SKILL.md).
Sanitization: No sanitization or escaping of the fetched titles is performed before they are used in the merge request creation command.
[EXTERNAL_DOWNLOADS]: The skill's metadata contains an update directive npx skills update and points to a source repository on GitHub. These resources are associated with the vendor's own namespace and infrastructure.

create-mr