Skills
Navigate
skills/autohandai/community-skills/analyzing-cloud-storage-access-patterns/Gen Agent Trust Hub

analyzing-cloud-storage-access-patterns

Pass

Audited by Gen Agent Trust Hub on Jul 16, 2026

Risk Level: SAFECOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [COMMAND_EXECUTION]: The script scripts/agent.py uses subprocess.run to call the aws CLI utility for querying CloudTrail events. Evidence:
  • The command is executed using a list of arguments, which prevents shell injection.
  • The operation is restricted to fetching metadata (lookup-events) from AWS.
  • [EXTERNAL_DOWNLOADS]: The skill instructions in SKILL.md require the installation of standard, well-known Python packages. Evidence:
  • pip install boto3 requests installs the official AWS SDK and a standard HTTP library.
  • [INDIRECT_PROMPT_INJECTION]: The skill processes external data (CloudTrail logs) that could technically be influenced by an attacker. Evidence:
  • Ingestion points: scripts/agent.py reads JSON output from the aws cloudtrail command.
  • Boundary markers: None present in the final JSON report generation.
  • Capability inventory: The script can execute subprocesses and write local files (open(args.output, "w")).
  • Sanitization: The script uses standard json.loads to parse data; it does not interpret the content as executable code or prompt instructions.
Audit Metadata
Risk Level
SAFE
Analyzed
Jul 16, 2026, 08:09 AM
Security Audit — agent-trust-hub — analyzing-cloud-storage-access-patterns