aws-monitoring

Pass

Audited by Gen Agent Trust Hub on May 12, 2026

Risk Level: SAFEPROMPT_INJECTION
Full Analysis
  • [PROMPT_INJECTION]: The skill provides reference commands for tailing and filtering logs from CloudWatch Logs (e.g., aws logs tail, aws logs filter-log-events). If the agent executes these commands and processes the resulting log data, it is exposed to an indirect prompt injection surface where log content could contain hidden instructions aimed at overriding agent behavior.
  • Ingestion points: SKILL.md facilitates reading untrusted external data via aws logs CLI commands.
  • Boundary markers: The provided command examples lack delimiters or explicit instructions to the agent to treat log content as data only and ignore any embedded instructions.
  • Capability inventory: The skill frontmatter allows the use of shell and write_file tools, which could be used to execute or store data extracted from logs.
  • Sanitization: No sanitization, filtering, or validation logic is present in the examples for handling the ingested log events.
Audit Metadata
Risk Level
SAFE
Analyzed
May 12, 2026, 05:07 PM
Security Audit — agent-trust-hub — aws-monitoring