The Agent Skills Directory

[INDIRECT_PROMPT_INJECTION]: The skill is designed to process external MSBuild binary log files (.binlog). While these files represent an attack surface for indirect prompt injection if they contain malicious diagnostic content, the risk is inherent to the tool's primary purpose of log analysis.
Ingestion points: .binlog files processed via the binlog MCP server tools or replayed using dotnet msbuild in SKILL.md.
Boundary markers: Absent; the agent is not specifically instructed to disregard instructions embedded in the logs.
Capability inventory: Build execution via dotnet msbuild and standard file system inspection using cat and grep as defined in SKILL.md.
Sanitization: None; the tool processes the binary data and generated text logs directly for debugging purposes.
[SAFE]: All operations and tools described in the skill, including the use of the binlog MCP server and the dotnet CLI, are standard development practices for MSBuild environments. No malicious command execution, data exfiltration, or obfuscation techniques were found.

binlog-failure-analysis