Skills
Navigate
skills/opsmill/infrahub-skills/infrahub-planning-upgrades/Gen Agent Trust Hub

infrahub-planning-upgrades

Pass

Audited by Gen Agent Trust Hub on Oct 8, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill ingests release notes from an external repository via the gh CLI. This content is untrusted and could potentially contain malicious instructions targeting the agent's behavior.
  • Ingestion points: Release note bodies fetched from the opsmill/infrahub repository.
  • Boundary markers: No explicit delimiters or boundary markers are used when processing these notes.
  • Capability inventory: Access to Bash for probes and Write for the upgrade plan.
  • Sanitization: No sanitization or validation of the fetched release note content is implemented.
  • [COMMAND_EXECUTION]: The skill uses Bash to run diagnostic probes, including infrahubctl commands and docker compose run for version-specific checks. These are restricted by safety rules to a predefined list of read-only probes.
  • [EXTERNAL_DOWNLOADS]: Fetches release information and technical notes from the official opsmill/infrahub repository on GitHub.
Audit Metadata
Risk Level
SAFE
Analyzed
Oct 8, 2026, 12:19 AM