pr-create
Pass
Audited by Gen Agent Trust Hub on May 7, 2026
Risk Level: SAFECOMMAND_EXECUTION
Full Analysis
- [COMMAND_EXECUTION]: The skill uses the
gitandgh(GitHub CLI) tools to manage local branches, stage changes, and create pull requests on GitHub. - [COMMAND_EXECUTION]: It dynamically identifies and executes local build, test, and lint commands by inspecting project files such as
package.json,Makefile, andpyproject.toml. - [INDIRECT_PROMPT_INJECTION]: The skill exposes an indirect prompt injection surface by ingesting and executing commands defined in local project configuration files.
- Ingestion points:
CLAUDE.md,AGENTS.md,package.json,Makefile,pyproject.toml, and.github/workflows/. - Boundary markers: None; the skill trusts the commands specified within these project files.
- Capability inventory: Shell execution of discovered scripts and full access to
gitandghCLI capabilities. - Sanitization: None; command strings found in configuration files are executed without validation or escaping.
Audit Metadata