MindWorkAI/AI-Studio
MindWorkAI/AI-Studio
1
0
Fork 0
mirror of https://github.com/MindWorkAI/AI-Studio.git synced 2026-03-22 22:11:36 +00:00
Code Issues Actions 1 Packages Projects Releases Wiki Activity

Updated agent instructions (#701)

Browse Source
This commit is contained in:
Thorsten Sommer 2026-03-21 13:55:37 +01:00 committed by GitHub
parent 070375c964
commit 86e2f1baac
No known key found for this signature in database
GPG Key ID: B5690EEEBB952194
1 changed files with 11 additions and 0 deletions
Show Stats Download Patch File Download Diff File Expand all files Collapse all files

11
AGENTS.md
View File

@ -29,6 +29,14 @@ dotnet run build
``` ```
This builds the .NET app as a Tauri "sidecar" binary, which is required even for development. This builds the .NET app as a Tauri "sidecar" binary, which is required even for development.
### Running .NET builds from an agent
- Do not run `.NET` builds such as `dotnet run build`, `dotnet build`, or similar build commands from an agent. Codex agents can hit a known sandbox issue during `.NET` builds, typically surfacing as `CSSM_ModuleLoad()` or other sandbox-related failures.
- Instead, ask the user to run the `.NET` build locally in their IDE and report the result back.
- Recommend the canonical repo build flow for the user: open an IDE terminal in the repository and run `cd app/Build && dotnet run build`.
- If the context fits better, it is also acceptable to ask the user to start the build using their IDE's built-in build action, as long as it is clear the build must be run locally by the user.
- After asking for the build, wait for the user's feedback before diagnosing issues, making follow-up changes, or suggesting the next step.
- Treat the user's build output, error messages, or success confirmation as the source of truth for further troubleshooting.
- For reference: https://github.com/openai/codex/issues/4915
### Running Tests ### Running Tests
Currently, no automated test suite exists in the repository. Currently, no automated test suite exists in the repository.
@ -177,12 +185,15 @@ Multi-level confidence scheme allows users to control which providers see which
## Important Development Notes ## Important Development Notes
- **File changes require Write/Edit tools** - Never use bash commands like `cat <<EOF` or `echo >` - **File changes require Write/Edit tools** - Never use bash commands like `cat <<EOF` or `echo >`
- **End of file formatting** - Do not append an extra empty line at the end of files.
- **Spaces in paths** - Always quote paths with spaces in bash commands - **Spaces in paths** - Always quote paths with spaces in bash commands
- **Agent-run .NET builds** - Do not run `.NET` builds from an agent. Ask the user to run the build locally in their IDE, preferably via `cd app/Build && dotnet run build` in an IDE terminal, then wait for their feedback before continuing.
- **Debug environment** - Reads `startup.env` file with IPC credentials - **Debug environment** - Reads `startup.env` file with IPC credentials
- **Production environment** - Runtime launches .NET sidecar with environment variables - **Production environment** - Runtime launches .NET sidecar with environment variables
- **MudBlazor** - Component library requires DI setup in Program.cs - **MudBlazor** - Component library requires DI setup in Program.cs
- **Encryption** - Initialized before Rust service is marked ready - **Encryption** - Initialized before Rust service is marked ready
- **Message Bus** - Singleton event bus for cross-component communication inside the .NET app - **Message Bus** - Singleton event bus for cross-component communication inside the .NET app
- **Empty lines** - Avoid adding extra empty lines at the end of files.
## Changelogs ## Changelogs
Changelogs are located in `app/MindWork AI Studio/wwwroot/changelog/` with filenames `vX.Y.Z.md`. These changelogs are meant to be for normal end-users Changelogs are located in `app/MindWork AI Studio/wwwroot/changelog/` with filenames `vX.Y.Z.md`. These changelogs are meant to be for normal end-users