agent-update-manager

Pass

Audited by Gen Agent Trust Hub on Aug 5, 2026

Risk Level: SAFEEXTERNAL_DOWNLOADS
Full Analysis
  • [SAFE]: The skill is purpose-built for repository management and version synchronization within the author's own ecosystem.
  • [EXTERNAL_DOWNLOADS]: The skill fetches version metadata and repository updates from 'https://github.com/lensetek/Digital-Marketing-Agent_Skills'. This source is owned by the skill author ('lensetek') and is used for its intended administrative purpose.
  • [PROMPT_INJECTION]: The skill processes remote data in the form of changelogs and version files, representing a potential indirect prompt injection surface.
  • Ingestion points: Remote 'plugin.json' and repository changelogs (SKILL.md workflow step 4).
  • Boundary markers: None explicitly defined in the workflow instructions.
  • Capability inventory: 'capabilities/file-operations' (read/write), 'capabilities/web-research' (network), 'capabilities/project-knowledge-graph'.
  • Sanitization: Not detailed in the workflow, but mitigated by the mandatory human-in-the-loop approval step (step 8) before any modifications are applied.
  • [DATA_EXFILTRATION]: The instructions include specific security guardrails to prevent the agent from exposing credentials, private remote URLs, or deployment secrets during the update process.
Audit Metadata
Risk Level
SAFE
Analyzed
Aug 5, 2026, 05:48 PM
Security Audit — agent-trust-hub — agent-update-manager