The Agent Skills Directory

[Command Execution]: The skill instructs the agent to perform basic shell checks to verify if a file exists before creating a new one. This is a standard and safe operational procedure.\n- [Data Access]: The agent reads configuration and source files within the repository to gather technical context, such as build scripts and code style. This access is localized and essential for generating accurate documentation.\n- [Indirect Prompt Injection Surface]: By analyzing content from external repositories, the skill includes a surface for indirect prompt injection. This is a typical consideration for tools that summarize external data, and the skill focuses on technical extraction to mitigate potential issues.

wiki-agents-md