gl-cli
Pass
Audited by Gen Agent Trust Hub on Oct 24, 2026
Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill instructs the agent to execute shell commands using the
zereight-mcp-gitlabCLI to fetch data and interact with GitLab repositories. - [EXTERNAL_DOWNLOADS]: The skill suggests installing the
@zereight/mcp-gitlabpackage from the npm registry. This package belongs to the skill's author. - [INDIRECT_PROMPT_INJECTION]: The skill facilitates the ingestion of untrusted data from GitLab (e.g., merge request descriptions, diffs, and discussions) which could contain malicious instructions.
- Ingestion points: Data is brought into the agent's context via the output of
zereight-mcp-gitlabcommands specified inSKILL.md. - Boundary markers: The instructions do not define specific delimiters or instructions to ignore potential commands within the fetched GitLab content.
- Capability inventory: The agent is capable of executing shell commands and installing packages.
- Sanitization: No explicit sanitization or filtering of the remote GitLab content is described in the skill instructions.
Audit Metadata