gpc-user-management
Warn
Audited by Gen Agent Trust Hub on Sep 26, 2026
Risk Level: MEDIUMEXTERNAL_DOWNLOADSINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTION
Full Analysis
- [EXTERNAL_DOWNLOADS]: The skill detection script (scripts/detect_gpc.mjs) and documentation (references/tester-workflows.md) reference 'npx gpc' and 'npx @gpc-cli/cli'. This causes the agent to download and execute third-party packages from the public NPM registry at runtime. These packages are not associated with a trusted organization or the verified vendor (yasserstudio).
- [INDIRECT_PROMPT_INJECTION]: The skill processes untrusted external data via CSV files for tester management, creating a vulnerability to indirect instructions.
- Ingestion points: The 'gpc testers import --file testers.csv' command documented in SKILL.md reads data from an external file into the agent's operational context.
- Boundary markers: The skill lacks instructions, delimiters, or 'ignore' warnings to isolate the imported data from the agent's core instructions.
- Capability inventory: The skill has high-privilege capabilities, including the ability to invite new users to the Google Play Console, modify account-level permissions, and manage per-app grants.
- Sanitization: There is no evidence of validation or sanitization for the content of the imported CSV files described in the procedures.
- [COMMAND_EXECUTION]: The scripts/detect_gpc.mjs utility executes shell commands via node:child_process (execSync) to check the environment, versioning, and authentication status of the GPC tool. These commands include 'gpc --version', 'npx gpc --version', and 'gpc auth status'.
Audit Metadata