glab-dependency-firewall

Pass

Audited by Gen Agent Trust Hub on Sep 17, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [COMMAND_EXECUTION]: The skill executes the glab command-line utility to interact with GitLab's Dependency Firewall experimental features. This is the core intended functionality provided by the skill.
  • [INDIRECT_PROMPT_INJECTION]: The skill acts as a surface for indirect prompt injection by processing untrusted data from Dependency Firewall logs.
  • Ingestion points: The ci-summary command reads and processes .gitlab/df/ci-log.json (referenced in SKILL.md and references/commands.md).
  • Boundary markers: The instructions do not define clear delimiters or "ignore instructions" wrappers for the summary data provided to the agent.
  • Capability inventory: The skill environment allows command execution (glab) and local file access within the workspace.
  • Sanitization: There is no mention of sanitization, filtering, or validation of the log content before it is presented to the agent for summarization.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 17, 2026, 08:57 AM
Security Audit — agent-trust-hub — glab-dependency-firewall