re-windbg
Pass
Audited by Gen Agent Trust Hub on Sep 14, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- [SAFE]: The skill provides instructions for downloading debugging information from Microsoft's official public symbol server at
msdl.microsoft.com. This is an established and trusted service for Windows diagnostics and development.- [INDIRECT_PROMPT_INJECTION]: The skill describes processes for analyzing external data sources such as crash dump files (.dmp) and Time Travel Debugging traces (.run). While these files are untrusted and could theoretically contain instructions to influence the agent, this is a necessary functionality for the tool's diagnostic purpose. - Ingestion points: The skill ingests untrusted data through the analysis of crash dumps, execution traces, and process memory attached during debugging sessions as noted in
SKILL.md. - Boundary markers: The provided instructions do not include boundary markers or delimiters to differentiate the untrusted data from the agent's operating context.
- Capability inventory: The skill uses WinDbg, which has extensive capabilities for system-level memory access, register modification, and the execution of scripts and commands.
- Sanitization: There are no procedures described in the skill for sanitizing the analyzed data or validating the integrity of the trace and dump files.
Audit Metadata