hcls-stack-check

Pass

Audited by Gen Agent Trust Hub on Nov 11, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONEXTERNAL_DOWNLOADS
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill processes data from external sources, specifically the AWS HIPAA Eligible Services list and AWS resource identifiers. This data is interpolated into the final report. Although the skill implements pipe-character escaping for markdown rendering and includes mandatory disclaimers and boundary markers to prevent the agent from misinterpreting data as instructions, the ingestion of untrusted external strings constitutes an indirect prompt injection surface.
  • Ingestion points: scripts/topology_source.py (resource ARNs from AWS APIs) and scripts/hipaa_source.py (scraping HTML from a public URL).
  • Boundary markers: scripts/report.py defines a mandatory BOUNDARY_STATEMENT and disclaimer text.
  • Capability inventory: Read-only AWS API access, network fetch via urllib, and markdown generation.
  • Sanitization: scripts/report.py uses _esc to handle markdown-breaking characters.
  • [COMMAND_EXECUTION]: The skill instructions in SKILL.md require the agent to execute a Python core (scripts/main.py) to handle inventory retrieval and eligibility matching. This execution is scoped to the skill's distributed scripts.
  • [EXTERNAL_DOWNLOADS]: The skill performs a live fetch of the HIPAA eligibility list from https://aws.amazon.com/compliance/hipaa-eligible-services-reference/. This is a trusted and well-known service domain essential for the skill's functionality.
Audit Metadata
Risk Level
SAFE
Analyzed
Nov 11, 2026, 02:01 PM