gpc-security
Pass
Audited by Gen Agent Trust Hub on Nov 2, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTIONEXTERNAL_DOWNLOADS
Full Analysis
- [COMMAND_EXECUTION]: The environment detection script (
scripts/detect_gpc.mjs) utilizesexecSyncto run informational commands likegpc --versionandgpc auth status. These commands are used solely to verify the installation and configuration of the GPC tool in the agent's environment. - [COMMAND_EXECUTION]: Rotation and cleanup procedures (
references/key-rotation.md) include shell commands forgcloudandgh(GitHub CLI) to manage cloud service account keys and repository secrets. It also includes targeted deletion commands for clearing the local token cache (rm -rf ~/.cache/gpc/tokens/) and secure file deletion (shred,rm -P). - [INDIRECT_PROMPT_INJECTION]: The skill documentation (
references/runtime-security.md) describes how the underlying GPC tool implements XML boundary tags to prevent indirect prompt injection during changelog generation from commit messages. This is documented as a security feature and mitigation strategy. - [EXTERNAL_DOWNLOADS]: The skill references the
yasserstudio/gpc-actionGitHub Action. This is a vendor-owned resource provided by the skill author for integrating GPC operations into GitHub Actions workflows.
Audit Metadata