The Agent Skills Directory

[COMMAND_EXECUTION]: The skill makes extensive use of shell commands to manage the CodeQL analysis pipeline. This includes running build systems (Make, CMake, Gradle, Maven, Cargo, dotnet), managing the CodeQL CLI, and using standard utilities (find, grep, awk, jq, fd, rg) for file system discovery and result processing. All commands are consistent with the skill's stated purpose of building and analyzing code databases.
[EXTERNAL_DOWNLOADS]: The skill facilitates the download of CodeQL query packs from trusted sources, specifically the official GitHub Security Lab and Trail of Bits repositories. It also manages the installation of target project dependencies using standard package managers (pip, npm, etc.) to ensure high-quality extraction. These operations are transparent and necessary for the intended functionality.
[DYNAMIC_EXECUTION]: The skill dynamically generates CodeQL query suites (.qls files) and diagnostic queries (.ql files) based on the detected programming language and the user's scan mode. This dynamic configuration is a standard technique for tailoring static analysis to the specific environment of the codebase being scanned.
[DATA_EXFILTRATION]: No evidence of unauthorized data exfiltration was found. The skill processes project source code and metadata locally to generate security findings and stores all artifacts within a user-defined or default output directory without transmitting sensitive information to external servers.

codeql