skills/aws-samples/sample-mainframe-modernization-agent-skills/fix-atx-datapath-errors/Gen Agent Trust Hub
fix-atx-datapath-errors
Pass
Audited by Gen Agent Trust Hub on Nov 11, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill defines a clear, multi-step workflow in SKILL.md that executes local Python scripts to parse, correlate, and fix mainframe source code issues. These operations are performed within the user's workspace using standard Python library modules like
zipfileandjson.- [INDIRECT_PROMPT_INJECTION]: The skill has a vulnerability surface for indirect prompt injection as it ingests untrusted data. Ingestion points:scripts/parse_debug_bundle.pyandscripts/correlate_with_source.pyread contents from user-provided ZIP files (JSON artifacts and COBOL/JCL source code). Boundary markers: No specific delimiters or safety instructions are used to separate untrusted data. Capability inventory: The skill involves file writing and command execution. Sanitization: Ingested content is not sanitized before being processed or included in reports. However, the risk is inherent to the skill's purpose and handled by the agent's internal safety guardrails.- [DATA_EXFILTRATION]: The skill does not perform any network operations. All activities are confined to reading frominput/and writing tooutput/directories. No access to sensitive system paths or credentials was detected.
Audit Metadata