glab-incident
Pass
Audited by Gen Agent Trust Hub on Nov 3, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests untrusted incident data (titles, descriptions, comments) from external GitLab projects, which provides an attack surface for malicious instructions to influence the agent. * Ingestion points: glab incident list and glab incident view commands (SKILL.md, references/commands.md). * Boundary markers: The instructions advise that the user or agent must "Review the complete body and attachments... before approving the post," but no technical delimiters are applied to the ingested content. * Capability inventory: The skill can perform write operations like posting notes and changing incident states via the glab CLI. * Sanitization: There is no mention of programmatic sanitization or filtering of the content retrieved from the GitLab API.
Audit Metadata