gitlab
Pass
Audited by Gen Agent Trust Hub on Mar 11, 2026
Risk Level: SAFEEXTERNAL_DOWNLOADSCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill requires the installation of the
@membranehq/clipackage via npm. This is a verified utility provided by the vendor (membrane) and is necessary for the skill to function. - [COMMAND_EXECUTION]: All GitLab operations are performed using the
membranecommand-line tool. The commands are limited to the intended functionality of managing GitLab projects, issues, and merge requests. - [INDIRECT_PROMPT_INJECTION]: The skill acts as an interface for GitLab data, which could potentially contain malicious content embedded in issue descriptions or project titles.
- Ingestion points: Data is retrieved from GitLab via actions such as
list-issuesandget-projectas defined inSKILL.md. - Boundary markers: None. There are no explicit instructions to the agent to disregard instructions found within the retrieved GitLab data.
- Capability inventory: The agent can perform actions and proxy requests using
membrane action runandmembrane request. - Sanitization: No sanitization or validation of the external GitLab content is specified.
Audit Metadata