The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes git commands and calls neighboring scripts context-audit/run.mjs and context-fix/run.mjs. These executions are limited to repo context maintenance and use standard Node.js execFileSync and spawnSync APIs.
[DATA_EXPOSURE_AND_EXFILTRATION]: The skill reads and writes JSON artifacts (plans and findings) strictly within the .git/context-spec directory of the target repository. It does not perform network operations or access sensitive credentials outside of standard Git repository metadata.
[REMOTE_CODE_EXECUTION]: The skill calls external scripts (context-audit and context-fix), but these are defined as mandatory separately-installed sibling skills within the local environment. There is no evidence of downloading or executing code from remote sources.
[DYNAMIC_EXECUTION]: The script uses spawnSync to invoke other local Node.js scripts using process.execPath. Path resolution is handled via standard Node.js path utilities, and inputs are validated against the repository root.

context-update