build-perf-diagnostics
Pass
Audited by Gen Agent Trust Hub on Nov 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTION
Full Analysis
- Indirect Prompt Injection Surface: The skill is designed to analyze performance data by reading build logs (
full.log) and binary logs (build.binlog) generated during the .NET build process. - Ingestion points: Data enters the agent's context through the reading of
full.logand binary logs via thebinlogMCP server orgrepcommands. - Boundary markers: The instructions do not specify the use of delimiters or warnings to ignore potentially malicious instructions embedded within the logs.
- Capability inventory: The skill utilizes shell-based tools (
dotnet,grep) to perform builds and search through generated diagnostic files. - Sanitization: There is no explicit sanitization or filtering of the log content before it is processed by the agent.
- Context: If a malicious project file or source code is crafted to emit specific strings during the build process, those strings could appear in the logs and influence the agent's subsequent analysis. This is a standard consideration for diagnostic tools that interpret output from external source code.
Audit Metadata