The Agent Skills Directory

[COMMAND_EXECUTION]: The skill requires the agent to run git commands, such as git rev-parse, git diff, and git log, to analyze the differences between release branches. This behavior is consistent with the skill's primary function of release auditing and operates within the expected development environment.
[INDIRECT_PROMPT_INJECTION]: The skill processes external data including git diffs, commit messages, and repository source code, which represents a potential attack surface for indirect prompt injection.
Ingestion points: Repository content and git history are accessed via git commands and repository-wide file searches in Step 3 and 4 of the workflow.
Boundary markers: The instructions do not specify explicit delimiters or warnings for the agent to ignore instructions that might be embedded within the analyzed diff or code content.
Capability inventory: The agent possesses the capability to read files, execute repository-level commands, and perform searches across the filesystem.
Sanitization: The skill does not define specific sanitization or filtering logic for the content it ingests from the repository history.

sdk-release