The Agent Skills Directory

[SAFE]: The skill implements a standard and well-documented workflow for diagnosing .NET Framework shim activation issues. It focuses on parsing human-readable log messages and HRESULTs to understand runtime resolution decisions.
[SAFE]: Command execution is limited to standard diagnostic tasks, such as using grep to search log files and PowerShell to read registry keys. These operations are used for information gathering and do not involve system modification or exfiltration.
[PROMPT_INJECTION]: The skill processes untrusted diagnostic log files (.CLRLoad*.log), creating a surface for indirect prompt injection. 1. Ingestion points: Data enters the agent context through the survey and analysis of log files. 2. Boundary markers: No explicit delimiters or instructions are used to separate log content from agent instructions. 3. Capability inventory: The skill uses grep for searching and PowerShell for registry inspection. 4. Sanitization: No validation or sanitization of the log content is performed before analysis. However, this is a theoretical surface common to diagnostic tools and is not considered a high-risk finding in this context.

clr-activation-debugging