7de87e9b3a
* The QA mode implementation is complete. Here's a summary of what was implemented:
## Summary
I've successfully implemented the QA mode feature for the vibe-kanban application. Here's what was created/modified:
### New Files
1. **`crates/services/src/services/qa_repos.rs`** - QA repos module that:
- Defines two hardcoded QA repo URLs (`internal-qa-1` and `internal-qa-2`)
- Clones repos to a persistent temp directory on first access using `git2`
- Returns `DirectoryEntry` compatible with existing API
2. **`crates/executors/src/executors/qa_mock.rs`** - Mock executor that:
- Implements `StandardCodingAgentExecutor` trait
- Performs random file operations (create, delete, modify)
- Streams 10 mock log entries in ClaudeJson format over 10 seconds
- Reuses Claude's log processor for normalization
### Modified Files
1. **Cargo.toml files** - Added `qa-mode` feature flags to:
- `crates/executors/Cargo.toml`
- `crates/services/Cargo.toml`
- `crates/server/Cargo.toml`
2. **`crates/executors/src/executors/mod.rs`** - Added:
- Conditional `qa_mock` module export
- `QaMock` variant to `CodingAgent` enum
- Updated `capabilities()` match arm
3. **`crates/executors/src/mcp_config.rs`** - Added `QaMock` handling
4. **`crates/services/src/services/filesystem.rs`** - Override `list_git_repos` and `list_common_git_repos` in QA mode
5. **`crates/services/src/services/container.rs`** - Override `normalize_logs` executor lookup in QA mode
6. **`crates/executors/src/actions/coding_agent_initial.rs`** - Override spawn in QA mode
7. **`crates/executors/src/actions/coding_agent_follow_up.rs`** - Override spawn in QA mode
8. **`vibe-kanban/package.json`** - Added QA build scripts:
- `backend:dev:qa`
- `backend:dev:watch:qa`
- `dev:qa`
### Usage
```bash
# Run backend in QA mode
cargo run --features qa-mode
# Or with pnpm
pnpm run dev:qa
```
When QA mode is enabled:
- `GitPanelCreate` will show only `internal-qa-1` and `internal-qa-2` as recent repos
- Starting any task will perform mock file operations and stream 10 logs over 10 seconds
- Logs follow the ClaudeJson format for compatibility with the existing log processor
* Cleanup script changes for workspace 15499061-ec87-4c1c-81b9-994e84d0e75b
* The TLS error fix is complete. I've replaced the `git2::Repository::clone` call with the `git` CLI approach in `qa_repos.rs`:
**Changes made:**
- Updated `clone_qa_repos_if_needed` function to use `Command::new("git")` with `["clone", "--depth", "1", url, path]` arguments instead of `git2::Repository::clone`
- Added `--depth 1` for shallow cloning (faster)
- Updated error handling to capture stderr output from the git command
The fix uses the git CLI which has reliable TLS support, avoiding the TLS configuration issues that git2 can have on some systems. This matches patterns used elsewhere in the codebase (e.g., `crates/remote/src/github_app/service.rs`).
You can now run `pnpm run backend:dev:qa` or `pnpm run dev:qa` and the QA repos should clone successfully.
* Cleanup script changes for workspace 15499061-ec87-4c1c-81b9-994e84d0e75b
* Done! I've made the mock logs strongly typed by:
1. **Added imports** for `ClaudeJson`, `ClaudeMessage`, `ClaudeContentItem`, and `ClaudeToolData` from the Claude executor
2. **Replaced `generate_mock_logs()`** to use strongly-typed structs:
- Each log entry is now a proper `ClaudeJson` enum variant
- Uses `ClaudeMessage` with proper fields (`id`, `message_type`, `role`, `model`, `content`, `stop_reason`)
- Uses `ClaudeContentItem` variants (`Thinking`, `ToolUse`, `ToolResult`, `Text`)
- Uses `ClaudeToolData` variants (`Read`, `Write`, `Bash`)
- Serializes to JSON using `serde_json::to_string()` for correct escaping
3. **Updated tests** to verify the logs deserialize back to `ClaudeJson` types
This ensures:
- All fields match what `ClaudeLogProcessor::process_logs()` expects
- No manual JSON escaping needed (serde handles it)
- Compile-time type checking prevents malformed logs
- Session IDs are consistent across all entries
* Cleanup script changes for workspace 15499061-ec87-4c1c-81b9-994e84d0e75b
* Done! I've fixed the log display issue by replacing the shell echo approach with a temp file approach:
**Changes made to `crates/executors/src/executors/qa_mock.rs`:**
1. **Write logs to a temp file** instead of embedding them in a shell script:
- Creates a temp file at `$TMPDIR/qa_mock_logs_{uuid}.jsonl`
- Writes all logs as newline-delimited JSON
2. **Use `IFS= read -r` to read the file** - this preserves exact content:
- `IFS=` prevents word splitting
- `-r` prevents backslash interpretation
- `echo "$line"` outputs the exact line content
3. **Clean up the temp file** after reading with `rm -f`
**Benefits:**
- No shell escaping needed - file contents are read verbatim
- Each line is output exactly as serialized by serde_json
- Avoids issues with single quotes, backslashes, and escape sequences
- Temp file is automatically cleaned up after use
This should fix the issue where logs were appearing as raw JSON or with corrupted content due to shell escaping issues.
* fix formatting
* Build succeeds. I've applied the same fix to the two remaining locations that had newlines in their content:
1. **Line 234** - README tool result: Changed `\n\n` to `\\n\\n`
2. **Line 302** - Bash tool result: Changed `\n` to `\\n`
This matches the pattern the user established in the assistant final message (line 318) where `\\n` is used instead of `\n` to ensure newlines are properly preserved through the serialization and display pipeline.
* Cleanup script changes for workspace 15499061-ec87-4c1c-81b9-994e84d0e75b
* simplify scripts
* update agents.md
Get 10X more out of Claude Code, Gemini CLI, Codex, Amp and other coding agents...
We're hiring!
Overview
AI coding agents are increasingly writing the world's code and human engineers now spend the majority of their time planning, reviewing, and orchestrating tasks. Vibe Kanban streamlines this process, enabling you to:
- Easily switch between different coding agents
- Orchestrate the execution of multiple coding agents in parallel or in sequence
- Quickly review work and start dev servers
- Track the status of tasks that your coding agents are working on
- Centralise configuration of coding agent MCP configs
- Open projects remotely via SSH when running Vibe Kanban on a remote server
You can watch a video overview here.
Installation
Make sure you have authenticated with your favourite coding agent. A full list of supported coding agents can be found in the docs. Then in your terminal run:
npx vibe-kanban
Documentation
Please head to the website for the latest documentation and user guides.
Support
We use GitHub Discussions for feature requests. Please open a discussion to create a feature request. For bugs please open an issue on this repo.
Contributing
We would prefer that ideas and changes are first raised with the core team via GitHub Discussions or Discord, where we can discuss implementation details and alignment with the existing roadmap. Please do not open PRs without first discussing your proposal with the team.
Development
Prerequisites
Additional development tools:
cargo install cargo-watch
cargo install sqlx-cli
Install dependencies:
pnpm i
Running the dev server
pnpm run dev
This will start the backend. A blank DB will be copied from the dev_assets_seed folder.
Building the frontend
To build just the frontend:
cd frontend
pnpm build
Build from source (macOS)
- Run
./local-build.sh - Test with
cd npx-cli && node bin/cli.js
Environment Variables
The following environment variables can be configured at build time or runtime:
| Variable | Type | Default | Description |
|---|---|---|---|
POSTHOG_API_KEY |
Build-time | Empty | PostHog analytics API key (disables analytics if empty) |
POSTHOG_API_ENDPOINT |
Build-time | Empty | PostHog analytics endpoint (disables analytics if empty) |
PORT |
Runtime | Auto-assign | Production: Server port. Dev: Frontend port (backend uses PORT+1) |
BACKEND_PORT |
Runtime | 0 (auto-assign) |
Backend server port (dev mode only, overrides PORT+1) |
FRONTEND_PORT |
Runtime | 3000 |
Frontend dev server port (dev mode only, overrides PORT) |
HOST |
Runtime | 127.0.0.1 |
Backend server host |
DISABLE_WORKTREE_ORPHAN_CLEANUP |
Runtime | Not set | Disable git worktree cleanup (for debugging) |
Build-time variables must be set when running pnpm run build. Runtime variables are read when the application starts.
Remote Deployment
When running Vibe Kanban on a remote server (e.g., via systemctl, Docker, or cloud hosting), you can configure your editor to open projects via SSH:
- Access via tunnel: Use Cloudflare Tunnel, ngrok, or similar to expose the web UI
- Configure remote SSH in Settings → Editor Integration:
- Set Remote SSH Host to your server hostname or IP
- Set Remote SSH User to your SSH username (optional)
- Prerequisites:
- SSH access from your local machine to the remote server
- SSH keys configured (passwordless authentication)
- VSCode Remote-SSH extension
When configured, the "Open in VSCode" buttons will generate URLs like vscode://vscode-remote/ssh-remote+user@host/path that open your local editor and connect to the remote server.
See the documentation for detailed setup instructions.