dependency-triage

Pass

Audited by Gen Agent Trust Hub on Jun 28, 2026

Risk Level: SAFE
Full Analysis
  • [SAFE]: The skill defines logical rules for classifying package updates (patch, minor, major) and handling CVEs. No obfuscation, malicious command execution, or unauthorized network activity was detected.
  • [INDIRECT_PROMPT_INJECTION]: The skill constitutes a surface for indirect prompt injection as it processes external, potentially untrusted data from package manifests and vulnerability databases to influence automated agent behavior.
  • Ingestion points: The skill instructions indicate the agent will scan package manifests, lockfiles, and CVE descriptions (SKILL.md).
  • Boundary markers: There are no explicit instructions to use boundary markers or to ignore instructions embedded within the ingested data.
  • Capability inventory: The skill is intended for use in a "dependency sweeper loop" with actions like "patch-in-worktree", which implies the agent has file system modification capabilities.
  • Sanitization: No sanitization or validation steps are defined for the data processed from external sources.
Audit Metadata
Risk Level
SAFE
Analyzed
Jun 28, 2026, 10:29 AM
Security Audit — agent-trust-hub — dependency-triage