github-pull-request
Pass
Audited by Gen Agent Trust Hub on Oct 7, 2026
Risk Level: SAFEINDIRECT_PROMPT_INJECTIONCOMMAND_EXECUTIONDYNAMIC_EXECUTION
Full Analysis
- [INDIRECT_PROMPT_INJECTION]: The skill ingests configuration from
.agents/pr-config.mdor.claude/pr-config.mdwithin the target repository. These files are used to override base branches, test commands, and checklist requirements. - Ingestion points: Repository-level configuration files (SKILL.md, Step 0).
- Boundary markers: None specified for separating repository-provided configuration from agent instructions.
- Capability inventory: Shell command execution via git, GitHub CLI, and test runners; repository modification via push and PR creation.
- Sanitization: No explicit validation or sanitization is described for the configuration values read from the repository.
- [COMMAND_EXECUTION]: The skill executes test commands inferred from common project files like
package.json,Makefile, andpyproject.toml, or as specified in the repository's own configuration. While expected for a development tool, this executes code defined within the repository being analyzed. - [DYNAMIC_EXECUTION]: The skill constructs and executes shell commands for
gitandghoperations using metadata extracted from the repository, such as branch names, commit messages, and file paths.
Audit Metadata