The Agent Skills Directory

[COMMAND_EXECUTION]: The skill executes various local shell commands to manage the development workflow. These include git for branch management and cherry-picking, gh (GitHub CLI) for interacting with pull requests, and make verify/make test for code generation and validation. These are standard operations for the skill's intended purpose in a Go-based project environment.
[INDIRECT_PROMPT_INJECTION]: The skill represents a surface for indirect prompt injection because it ingests data from external pull requests that is later used in command arguments.
Ingestion points: Pull request titles, bodies, and commit messages are fetched via gh pr view in SKILL.md.
Boundary markers: The skill does not explicitly define sanitization or escaping for the interpolated PR content when creating the new PR.
Capability inventory: The agent has the ability to execute shell commands (git, gh, make) and write to the file system.
Sanitization: No specific sanitization logic is implemented for the external strings before they are used in the gh pr create command, though the risk is mitigated by the is_bot: true validation which limits processing to automated accounts.

fix-hypershift-repo-robot-pr