land
Pass
Audited by Gen Agent Trust Hub on Nov 22, 2026
Risk Level: SAFECOMMAND_EXECUTIONINDIRECT_PROMPT_INJECTION
Full Analysis
- [COMMAND_EXECUTION]: The skill automates repository workflows by executing shell commands. It uses the GitHub CLI (
gh) for API operations and Git (git) for managing branches and commits. It also runs local Node.js scripts, includingland.mjsfor PR logic andboard.mjsfor project management. - [INDIRECT_PROMPT_INJECTION]: The skill analyzes Pull Request titles and descriptions to identify linked issues and determine automation steps. This constitutes an ingestion of data from external contributors.
- Ingestion points: Pull Request body and title data are read in
land.mjsand used to identify issue closing references. - Boundary markers: The skill does not use explicit delimiters for PR content, relying instead on regex patterns to extract relevant identifiers.
- Capability inventory: The skill possesses the ability to merge branches, update project boards, and execute various repository-level commands via
ghandgit. - Sanitization: PR content is parsed for specific identifiers rather than being interpreted as direct commands.
Audit Metadata