xmake-trybuild

Pass

Audited by Gen Agent Trust Hub on Sep 21, 2026

Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
  • [INDIRECT_PROMPT_INJECTION]: The skill is designed to process and build third-party projects that contain build configuration files such as CMakeLists.txt, configure scripts, or Makefile. This creates a vulnerability surface where a malicious project could embed instructions within these files or their resulting build output to influence the agent's behavior.
  • Ingestion points: The skill interacts with the local file system to detect and read build configuration files in third-party project roots.
  • Boundary markers: No explicit boundary markers or instructions to ignore embedded content are provided.
  • Capability inventory: The skill utilizes shell command execution to invoke xmake and its sub-processes (cmake, make, meson, etc.).
  • Sanitization: No sanitization of third-party build scripts or their output is mentioned.
  • [COMMAND_EXECUTION]: The skill instructions facilitate the execution of shell commands to drive build backends. It specifically notes the use of the -y flag (e.g., xmake -y) to skip confirmation prompts, which increases the level of autonomy and reduces user oversight during the build process.
Audit Metadata
Risk Level
SAFE
Analyzed
Sep 21, 2026, 12:42 PM
Security Audit — agent-trust-hub — xmake-trybuild