.readignore/.writeignore

For cybersecurity reasons, it may be useful to prevent the agent from reading or editing certain folders or files in the project. This can be done by creating .readignore and .writeignore files in the .explyt/tools/ folder in the project.

The files support a subset of the .gitignore syntax: the *, ? and ** patterns, comments, negations, and the "last match wins" rule. Character classes such as [a-z] are not supported: square brackets are treated as ordinary characters.

If you add a file or folder to .writeignore, the agent will be able to read it but won't be able to edit it. This can be useful, for example, if you want to change your code in a TDD style with the agent, but ensure that tests won't be adjusted to fit the implementation.

If you add a file or folder to .readignore, the agent won't be able to read it. .readignore does not forbid writing: the write check uses .writeignore only. To block the agent's access to a path completely, add it to both files. This is useful if a file contains confidential data, such as API keys for paid services that the application requires but the agent should not see.

File access outside the project

The agent can read and write files outside the project root. By default, it is limited to the project directory and a few allowed paths.

To allow access to external files, open Settings → Tools → Explyt → Agent → File Access.

This is useful for projects that reference external files: shared configs, monorepos with sibling projects, or data files kept outside the working directory.