The Agent Skills Directory

[COMMAND_EXECUTION]: The skill uses git log to determine the last time the README was updated and to categorize changes since that date. These are standard read-only operations for repository metadata.
[DATA_EXFILTRATION]: No network-capable tools are utilized, and there are no attempts to access sensitive files outside of the project directory.
[PROMPT_INJECTION]: Instructions are focused on documentation maintenance and do not contain patterns attempting to bypass safety protocols or override agent behavior.
[INDIRECT_PROMPT_INJECTION]: The skill is designed to ingest data from untrusted sources within the project (such as existing README content and manifest files). While this represents a potential attack surface for indirect prompt injection, it is inherent to the skill's primary function of documentation synthesis. No markers or sanitization logic are explicitly defined, but the risk is considered low given the context of maintaining local project documentation.

update-readme