* The implementation is complete. Here's a summary of what was done:
Successfully added environment variable configuration support to executor profiles. Users can now configure env vars like this:
```json
{
"CLAUDE_CODE": {
"DEFAULT": {
"CLAUDE_CODE": {
"env": {
"ANTHROPIC_AUTH_TOKEN": "...",
"ANTHROPIC_BASE_URL": "https://api.z.ai/api/anthropic",
"API_TIMEOUT_MS": "3000000",
"ANTHROPIC_DEFAULT_SONNET_MODEL": "glm-4.6"
}
}
}
}
}
```
1. **`crates/executors/src/command.rs`**:
- Added `env: Option<HashMap<String, String>>` to `CmdOverrides` struct
- Added `apply_env_vars()` helper function
2. **Updated all executors to apply env vars**:
- `claude.rs` - `spawn_internal()`
- `codex.rs` - `spawn()`
- `amp.rs` - `spawn()` and `spawn_follow_up()`
- `droid.rs` - module-level `spawn()` function
- `copilot.rs` - both spawn methods
- `cursor.rs` - both spawn methods
- `opencode.rs` - both spawn methods
- `acp/harness.rs` - `spawn_with_command()` and `spawn_follow_up_with_command()` (used by Gemini/Qwen)
3. **Type Generation**: All JSON schemas and TypeScript types now include the `env` field
This allows users to set environment variables for all supported coding agents (Claude Code, Codex, Gemini, Amp, Droid, Copilot, Cursor, OpenCode, Qwen) through profile configuration, eliminating the need for external tools like claude-code-router.
* The implementation is complete. Here's a summary of what was done:
## Summary
### Backend Changes
- Added `env: Option<HashMap<String, String>>` field to `CmdOverrides` struct in `crates/executors/src/command.rs`
- Created `apply_env_vars()` helper function to inject env vars into `tokio::process::Command`
- Updated all 9 executors to apply env vars during spawn:
- claude.rs, codex.rs, amp.rs, droid.rs, copilot.rs, cursor.rs, opencode.rs
- gemini.rs and qwen.rs (via ACP harness)
- Modified ACP harness signature to accept `Option<&CmdOverrides>`
### Frontend Changes
- Created `KeyValueField.tsx` - custom RJSF field for editing key-value pairs
- Registered the field in `theme.ts`
- Added `uiSchema` to `ExecutorConfigForm.tsx` to use the custom field for `env`
### Generated Files
- `shared/types.ts` - TypeScript types updated with `env` field
- `shared/schemas/*.json` - All 9 executor schemas include `env` property
The environment variables UI will now appear in Settings > Agent Settings as a key-value editor with "Environment Variables" label and description. Users can add/remove/edit env vars that will be passed to the CLI execution environment.
* cleanup env structs
* fix form
* fmt
* remove mise.toml
* fmt
* Seprate config form per selected variant
---------
Co-authored-by: Louis Knight-Webb <louis@bloop.ai>
Co-authored-by: Solomon <abcpro11051@disroot.org>
* The implementation is complete. Here's a summary of what was done:
I've implemented environment variable injection when launching coding agents. The following environment variables are now injected into the shell when an executor is spawned:
- `VK_PROJECT_NAME` - Name of the project
- `VK_TASK_ID` - UUID of the task
- `VK_ATTEMPT_ID` - UUID of the task attempt
- `VK_ATTEMPT_BRANCH` - Branch name for the attempt (e.g., `vk/branch-name`)
1. **Created `crates/executors/src/env.rs`** - New module containing `ExecutionEnv` struct with:
- `HashMap<String, String>` to store env vars
- `insert()` method to add env vars
- `apply_to_command()` method to apply env vars to a `tokio::process::Command`
2. **Updated `crates/executors/src/actions/mod.rs`** - Modified `Executable` trait to accept `Option<&ExecutionEnv>` parameter
3. **Updated `crates/executors/src/executors/mod.rs`** - Modified `StandardCodingAgentExecutor` trait's `spawn()` and `spawn_follow_up()` to accept `Option<&ExecutionEnv>` parameter
4. **Updated all action implementations**:
- `coding_agent_initial.rs` - forwards env to agent spawn
- `coding_agent_follow_up.rs` - forwards env to agent spawn
- `script.rs` - applies env vars to script commands
5. **Updated all executor implementations** to apply env vars:
- `claude.rs`, `amp.rs`, `gemini.rs`, `codex.rs`, `qwen.rs`, `droid.rs`, `cursor.rs`, `copilot.rs`, `opencode.rs`
- Also updated `acp/harness.rs` which is used by gemini and qwen
6. **Updated `crates/local-deployment/src/container.rs`** - Modified `start_execution_inner()` to:
- Load task and project context from database
- Create `ExecutionEnv` with the four `VK_*` variables
- Pass the env to `executor_action.spawn()`
* Cleanup script changes for task attempt 4dc9b26c-c637-465a-ae71-644c828ea8f7
* The refactoring is complete. I've successfully updated the `env` parameter from `Option<&ExecutionEnv>` to `&ExecutionEnv` across all relevant files:
**Summary of changes:**
1. **Trait updates:**
- `Executable` trait in `crates/executors/src/actions/mod.rs`
- `StandardCodingAgentExecutor` trait in `crates/executors/src/executors/mod.rs`
2. **Action implementations:**
- `coding_agent_initial.rs`
- `coding_agent_follow_up.rs`
- `script.rs`
3. **Executor implementations (all 10 executors):**
- `claude.rs` - spawn, spawn_follow_up, spawn_internal
- `amp.rs` - spawn, spawn_follow_up
- `gemini.rs` - spawn, spawn_follow_up
- `codex.rs` - spawn, spawn_follow_up, spawn_inner
- `qwen.rs` - spawn, spawn_follow_up
- `droid.rs` - spawn, spawn_follow_up, spawn_droid helper
- `cursor.rs` - spawn, spawn_follow_up
- `copilot.rs` - spawn, spawn_follow_up
- `opencode.rs` - spawn, spawn_follow_up
- `acp/harness.rs` - spawn_with_command, spawn_follow_up_with_command
4. **Call site:**
- `container.rs` in `start_execution_inner` - changed `Some(&env)` to `&env`
All tests pass (137 tests) and clippy reports no warnings.
* All tests pass. The change is complete.
**Summary**: Added `VK_PROJECT_ID` to the environment variables injected when launching coding agents.
The environment variables now include:
- `VK_PROJECT_NAME` - project name
- `VK_PROJECT_ID` - project UUID (newly added)
- `VK_TASK_ID` - task UUID
- `VK_ATTEMPT_ID` - attempt UUID
- `VK_ATTEMPT_BRANCH` - git branch name for the attempt
* undo ts-rs bindings
* All 147 tests pass. The rebase conflict has been resolved successfully.
**Summary:**
- Resolved the rebase conflict in `crates/local-deployment/src/container.rs`
- The conflict was in the imports section where:
- `origin/main` added `InterruptSender` to the executors import
- Our branch added `env::ExecutionEnv` import
- Resolution: Combined both changes to keep `ExecutionEnv` import AND `InterruptSender`
- All tests pass (147 tests)
The branch `vk/4dc9-inject-env-vars` is now rebased onto `origin/main` with the complete environment variable injection feature, including:
- `VK_PROJECT_NAME`
- `VK_PROJECT_ID`
- `VK_TASK_ID`
- `VK_ATTEMPT_ID`
- `VK_ATTEMPT_BRANCH`
* remove bindings (again)
* I have bumped the `amp` version to `0.0.1764705684-g95eb77` in [crates/executors/src/executors/amp.rs](file:///private/var/folders/fr/0c4ky3392mb4yz5knw_wjdd00000gn/T/vibe-kanban/worktrees/5070-bump-amp/crates/executors/src/executors/amp.rs).
I verified the changes by running `cargo check --workspace`, which passed successfully.
* I have updated the `amp` version to `0.0.1764777697-g907e30` in [crates/executors/src/executors/amp.rs](file:///private/var/folders/fr/0c4ky3392mb4yz5knw_wjdd00000gn/T/vibe-kanban/worktrees/5070-bump-amp/crates/executors/src/executors/amp.rs).
I verified the changes by running `cargo check --workspace` (after resolving a transient `libsqlite3-sys` build issue).
* fix: No conversation found with session ID issue
Load session id after session is initialized.
* fmt
* claude: ignore session from message fragments
* fix tests
* feat: Add setting to use Claude subscription when API key is detected
This commit adds a new optional setting `use_claude_subscription` to the ClaudeCode
executor configuration. When enabled, this setting removes the ANTHROPIC_API_KEY
environment variable before spawning the Claude Code agent, ensuring that users with
Claude Pro/Team subscriptions can opt to use their subscription instead of being
charged API fees.
## Changes
- Added `use_claude_subscription` optional field to the `ClaudeCode` struct
- Implemented logic in `spawn_internal` to conditionally remove `ANTHROPIC_API_KEY`
from the environment when the setting is enabled
- Added tracing log when API key is removed for better debugging
## Implementation Details
- The field is optional (`Option<bool>`) and defaults to `false` when not set,
maintaining backward compatibility
- Uses `#[serde(skip_serializing_if = "Option::is_none")]` to keep JSON clean
- The setting is automatically exposed in the frontend via the JSON Schema
auto-generation from Rust structs
- TypeScript bindings are auto-generated via the `#[derive(TS)]` macro
## Benefits
- Prevents unexpected API charges for users with Claude subscriptions
- Gives users explicit control over authentication method
- Backward compatible - existing configurations continue to work unchanged
- No frontend changes needed - the setting appears automatically in the
ExecutorConfigForm
## Related
- Addresses feature request in discussion #1228
- Design document: https://github.com/potable-anarchy/vibe-kanban-launcher/blob/main/DESIGN_PROPOSAL.md
* cleanups
rename the config parameter to `disable_api_key`.
regenerate type bindings.
* suggest using the setting in the api-key warning
---------
Co-authored-by: Solomon <abcpro11051@disroot.org>
* Done. Updated Amp version to `0.0.1764081384-g1961a83` in [crates/executors/src/executors/amp.rs](file:///private/var/folders/m1/9q_ct1913z10v6wbnv54j25r0000gn/T/vibe-kanban/worktrees/ba61-bump-amp/crates/executors/src/executors/amp.rs#L36).
* Updated to `0.0.1764081384-g1961a8`.
* Add codex setup helper
Pass exit result
Move codex setup to routes
FIx fmt
Fix finalize
* Rename scriptcontext (vibe-kanban 79207902)
Rename the gh cli script context to something more general and use it for installs in crates/server/src/routes/task_attempts/cursor_setup.rs
Rename scriptcontext (vibe-kanban 79207902)
Rename the gh cli script context to something more general and use it for installs in crates/server/src/routes/task_attempts/cursor_setup.rs
Fmt
* Fix missing overrides for codex
* Perfect! All the implementation is complete. Let me summarize what was done:
I've successfully implemented agent availability checking during onboarding, mirroring the existing IDE availability functionality. Here's what was added:
1. **New API endpoint** in `crates/server/src/routes/config.rs`:
- Added `CheckAgentAvailabilityQuery` struct with `executor: BaseCodingAgent`
- Added `CheckAgentAvailabilityResponse` struct with `available: bool`
- Added `check_agent_availability()` handler that uses the existing `check_availability()` method from the executors crate
- Registered route: `/api/agents/check-availability`
2. **Type generation** in `crates/server/src/bin/generate_types.rs`:
- Added the new types to the TypeScript generation list
3. **API client** in `frontend/src/lib/api.ts`:
- Added `checkAgentAvailability()` method to `configApi`
- Imported necessary types
4. **Custom hook** in `frontend/src/hooks/useAgentAvailability.ts`:
- Created hook that checks agent availability asynchronously
- Returns states: `'checking' | 'available' | 'unavailable' | null`
5. **UI component** in `frontend/src/components/AgentAvailabilityIndicator.tsx`:
- Visual indicator showing spinner (checking), checkmark (available), or warning icon (unavailable)
- Matches the style of `EditorAvailabilityIndicator`
6. **Onboarding dialog** in `frontend/src/components/dialogs/global/OnboardingDialog.tsx`:
- Added agent availability check using the hook
- Displays the availability indicator below the agent selection dropdown
- The backend leverages existing `check_availability()` method that checks for MCP config files
- Frontend pattern exactly mirrors the IDE availability checking for consistency
- All TypeScript types are auto-generated from Rust
- Both backend (Rust) and frontend (TypeScript) checks pass successfully
The implementation is complete and ready for use!
Cleanup script changes for task attempt 60b68945-b59c-46e1-895d-540a910f2904
Great! TypeScript compilation passes. The semantic color classes have been applied successfully. The component now uses:
- `text-success` for available agents (instead of `text-green-500`/`text-green-600`)
- `text-warning` for unavailable agents (instead of `text-orange-500`/`text-orange-600`)
This matches the semantic color conventions used throughout the codebase and will properly adapt to theme changes.
Excellent! I've implemented file modification time checking for Claude Code's credential file. Here's what the enhanced availability check does:
The `check_availability()` method for Claude Code now validates:
1. **File exists**: `~/.claude.json` must exist
2. **File is fresh**: Credentials file must have been modified within the last **60 days**
- If the file hasn't been touched in 60 days, it's considered stale and the agent is marked as unavailable
- This handles the case where credentials might have expired but the file still exists
- **OAuth refresh updates the file**: When credentials are refreshed (e.g., access token renewed), the file gets rewritten with new tokens, updating the modification time
- **Simple and fast**: Just a filesystem metadata check, no JSON parsing or API calls needed
- **Conservative threshold**: 60 days is reasonable - most OAuth refresh tokens are valid for 30-90 days, so if nothing has touched the file in 60 days, it's likely expired
- **False negatives possible**: If someone logged in 61 days ago and never used Claude Code since, it might still work (refresh token could be valid for 90 days), but we'll show it as unavailable
- **False positives possible**: File could be recently modified but tokens could be manually revoked server-side
- **Not foolproof**: The only way to be 100% certain is to make an authenticated API call
But as a heuristic for showing availability during onboarding, this is a solid approach that balances accuracy with performance.
Cleanup script changes for task attempt 60b68945-b59c-46e1-895d-540a910f2904
Perfect! All checks pass. Let me summarize what was implemented:
- Enhanced `check_availability()` to check `~/.claude.json` modification time
- Marks agent as unavailable if credential file is older than 60 days
- Uses `std::fs::metadata()` and `modified().elapsed()` to compute age
- `CheckAgentAvailabilityResponse` now includes `credential_last_modified: Option<i64>`
- Calls new trait method `get_credential_last_modified()` to get Unix timestamp
- Returns `None` if file doesn't exist or metadata can't be read
- Added `get_credential_last_modified()` to `StandardCodingAgentExecutor` trait
- Gets file metadata and converts modification time to Unix timestamp
- Works for all agents automatically (uses their `default_mcp_config_path()`)
- On mount, checks availability of ALL agents in parallel
- Filters to only available agents with credentials
- Sorts by most recent `credential_last_modified` timestamp
- Auto-selects the agent with freshest credentials
- Only runs if user hasn't already configured an agent
1. User opens onboarding dialog for the first time
2. Frontend checks all agents in parallel (CLAUDE_CODE, CODEX, CURSOR_AGENT, etc.)
3. Each returns `{ available: bool, credential_last_modified: timestamp }`
4. Frontend finds agent with most recent credential file modification
5. Auto-selects that agent (likely the one they used most recently)
6. User can still manually change if they want
This provides a smart default while still allowing user choice!
Perfect! All checks pass. Let me show you the final summary:
I've successfully implemented comprehensive agent availability checking with automatic selection of the most recently used agent. Here's what was added:
**`AvailabilityInfo` struct** (`crates/executors/src/executors/mod.rs:167-206`):
- Structured type with fields: `mcp_config_found`, `auth_config_found`, `auth_last_edited`
- `is_available()` method that determines availability from the signals
- Logic: Available if MCP config exists OR auth config exists and is < 60 days old
**Trait method** (`crates/executors/src/executors/mod.rs:229-241`):
- Added `get_availability_info() -> AvailabilityInfo` to `StandardCodingAgentExecutor`
- Default implementation checks only MCP config
- Updated `check_availability()` to use `get_availability_info().is_available()`
**Claude Code** (`crates/executors/src/executors/claude.rs:200-224`):
- Auth credentials: `~/.claude.json`
- Checks file existence and modification time
- Returns AvailabilityInfo with auth signals
**Codex** (`crates/executors/src/executors/codex.rs:168-195`):
- MCP config: `~/.codex/config.toml`
- Auth credentials: `~/.codex/auth.json`
- Checks both MCP and auth configs
**Gemini** (`crates/executors/src/executors/gemini.rs:96-123`):
- MCP config: `~/.gemini/settings.json`
- Auth credentials: `~/.gemini/oauth_creds.json`
- Checks both MCP and auth configs
**Backend API** (`crates/server/src/routes/config.rs:479-498`):
- Returns `credential_last_modified` timestamp from AvailabilityInfo
- Uses `info.is_available()` for the available field
**Frontend auto-selection** (`frontend/src/components/dialogs/global/OnboardingDialog.tsx:64-112`):
- On mount, checks all agents in parallel
- Filters to available agents with credentials
- **Sorts by most recent `credential_last_modified`**
- Auto-selects the agent used most recently
This provides intelligent defaults while respecting user choice!
Perfect! All checks pass. Here's what I implemented:
**Enhanced API Response** (`crates/server/src/routes/config.rs:472-481`):
```rust
pub struct CheckAgentAvailabilityResponse {
available: bool,
mcp_config_found: bool,
auth_config_found: bool,
credential_last_modified: Option<i64>,
}
```
Now returns all the availability signals from `AvailabilityInfo`.
**OnboardingDialog** (`frontend/src/components/dialogs/global/OnboardingDialog.tsx`):
1. **State for availability map** (lines 57-59):
- Stores full availability info for all agents
- Used for intelligent sorting
2. **Priority-based sorting** (lines 187-218):
- **Tier 1**: Agents with auth config - sorted by most recent first
- **Tier 2**: Agents with MCP config only - alphabetical within tier
- **Tier 3**: Other agents - alphabetical within tier
- **Fallback**: Alphabetical if no availability info loaded yet
```typescript
// Tier 1: Both have auth - sort by recency
if (infoA.auth_config_found && infoB.auth_config_found) {
return timeB - timeA; // Most recent first
}
// Tier 2: Auth trumps MCP-only
if (infoA.auth_config_found) return -1;
if (infoB.auth_config_found) return 1;
// Tier 3: Both have MCP only - alphabetical
if (infoA.mcp_config_found && infoB.mcp_config_found) {
return a.localeCompare(b);
}
```
The dropdown will now show agents in this order:
1. **Most recently authenticated agents first** (e.g., Claude Code used yesterday)
2. **Other authenticated agents** (e.g., Gemini used last week)
3. **Agents with only MCP config** (alphabetical)
4. **Agents with no config** (alphabetical)
This ensures users see their most relevant agents at the top of the list!
Cleanup script changes for task attempt 60b68945-b59c-46e1-895d-540a910f2904
Perfect! All checks pass. Here's the final implementation summary:
**Removed `auth_config_found` field** - simplified to just use `auth_last_edited.is_some()`:
**AvailabilityInfo** (`crates/executors/src/executors/mod.rs:167-201`):
```rust
pub struct AvailabilityInfo {
pub mcp_config_found: bool,
pub auth_last_edited: Option<i64>, // If Some(_), auth exists
}
```
**Implementations:**
- **Claude Code** (`claude.rs:200-212`): `~/.claude.json` → `auth_last_edited`
- **Codex** (`codex.rs:168-185`): MCP at `~/.codex/config.toml`, auth at `~/.codex/auth.json`
- **Gemini** (`gemini.rs:96-113`): MCP at `~/.gemini/settings.json`, auth at `~/.gemini/oauth_creds.json`
**CheckAgentAvailabilityResponse** (`config.rs:472-480`):
```rust
{
available: bool,
mcp_config_found: bool,
credential_last_modified: Option<i64>, // If Some, auth file exists
}
```
**Priority-based dropdown ordering** (`OnboardingDialog.tsx:191-230`):
1. **Tier 1**: Agents with auth (`credential_last_modified !== null`) - sorted by most recent
2. **Tier 2**: Agents with MCP only - alphabetical
3. **Tier 3**: Others - alphabetical
Users will see:
- **Top of list**: Their most recently authenticated agents (e.g., "logged into Claude Code yesterday")
- **Middle**: Agents with MCP config but no recent auth
- **Bottom**: Remaining agents alphabetically
The implementation is clean, simple, and provides intelligent defaults!
Cleanup script changes for task attempt 60b68945-b59c-46e1-895d-540a910f2904
Perfect! All checks pass. The fix is complete:
**The Problem:**
- Line 76 had: `if (!profiles || config?.executor_profile) return;`
- This meant availability was **never fetched** for users with existing config
- `agentAvailabilityMap` stayed empty, sorting defaulted to alphabetical
- Result: AMP appeared first
**The Solution:**
- Removed `config?.executor_profile` from the skip condition
- Now availability is **always fetched** to enable sorting
- Moved the `config?.executor_profile` check **inside** the function (line 116)
- Auto-selection only happens for new users, but sorting works for everyone
**Now the dropdown will show:**
1. **Agents with recent auth** - sorted by most recent first (e.g., Claude Code used yesterday)
2. **Agents with MCP config only** - alphabetical (e.g., Cursor with MCP setup)
3. **Other agents** - alphabetical
The priority-based sorting is now fully functional!
Perfect! Fixed. Both indicators now use the same semantic colors:
- `text-success` for available (instead of `text-green-500`/`text-green-600`)
- `text-warning` for unavailable (instead of `text-orange-500`/`text-orange-600`)
The colors will now match perfectly between IDE and coding agent availability indicators.
Great! TypeScript compiles successfully. The changes are complete:
**AgentAvailabilityIndicator** now has:
1. **Updated text:**
- "Available" → "Login Detected"
- "Not Found" remains "Not Found"
2. **Tooltips on hover:**
- **Login Detected**: "Recent authentication credentials found for this agent"
- **Not Found**: "No credentials detected. The agent may still work if configured differently."
3. **Visual improvements:**
- Added `cursor-help` class to show the help cursor on hover
- Wrapped with shadcn/ui `Tooltip` components
This provides clearer messaging and helpful context for users!
Perfect! The tooltips now have `className="z-[100]"` which should render them above the modal. The fix ensures the hover text is readable when shown within the onboarding dialog.
Perfect! The tooltips now use `z-[10000]` which is higher than the modal's `z-[9999]`, so they should appear on top and be readable.
Perfect! TypeScript compiles successfully. The agent availability indicator is now added to the AgentSettings page:
**AgentSettings.tsx** now has:
1. **Availability check** (line 65):
```typescript
const agentAvailability = useAgentAvailability(selectedExecutorType);
```
2. **Indicator display** (line 444):
```typescript
<AgentAvailabilityIndicator availability={agentAvailability} />
```
The indicator appears below the agent dropdown selector and shows:
- "Login Detected" with tooltip for agents with credentials
- "Not Found" with tooltip for agents without credentials
- Same styling and behavior as the onboarding dialog
This provides consistent availability feedback throughout the app!
Cleanup script changes for task attempt 60b68945-b59c-46e1-895d-540a910f2904
Excellent! TypeScript compiles successfully. The changes are complete:
**Reverted AgentSettings.tsx:**
- Removed the availability indicator from the Agent Settings page (which is for advanced profile configuration)
**Added to GeneralSettings.tsx:**
- **Line 43-44**: Imported `useAgentAvailability` hook and `AgentAvailabilityIndicator` component
- **Line 81**: Added hook call: `const agentAvailability = useAgentAvailability(draft?.executor_profile?.executor);`
- **Line 420**: Added indicator display: `<AgentAvailabilityIndicator availability={agentAvailability} />`
The agent availability indicator now appears in the **General Settings** page under the "Task Execution" section, right below the agent dropdown selector - matching the pattern of how editor availability is shown in the same settings page. This is the correct location since it's where users configure their default agent, not the advanced agent configuration page.
Cleanup script changes for task attempt 60b68945-b59c-46e1-895d-540a910f2904
Perfect! All checks pass. Comments removed and types regenerated successfully. The implementation is complete and clean.
Fix agent availability i18n (vibe-kanban dcd1fc06)
Run @scripts/check-i18n.sh until it passes. Make sure to check the script and set GITHUB_BASE_REF to main
Fix agent availability i18n (vibe-kanban 77795175)
Run @scripts/check-i18n.sh until it passes. Make sure to check the script and set GITHUB_BASE_REF to main
Currently agents.availability.loginDetected shows in the frontend so it's not using i18n correctly, it was added in the last commit though
Perfect! TypeScript compiles. Now the availability indicator shows three distinct states:
1. **"Login Detected"** - Has auth credentials (recent login)
2. **"Installation Found"** - Has MCP config but no auth (agent is installed)
3. **"Not Found"** - Neither auth nor MCP config found
This gives users much clearer information about what was actually detected!
Cleanup script changes for task attempt 60b68945-b59c-46e1-895d-540a910f2904
Perfect! Now the indicator shows three distinct states:
1. **"Login Detected"** (green check) - Has recent auth credentials
- Tooltip: "Recent authentication credentials found for this agent"
2. **"Installation Found"** (green check) - Has MCP config but no auth
- Tooltip: "Agent configuration found. You may need to log in to use it."
3. **"Not Found"** (orange warning) - Neither found
- Tooltip: "No credentials detected. The agent may still work if configured differently."
This gives users accurate, specific information about what was actually detected!
Fix i18n agent availability (vibe-kanban 9eab4ea1)
Run @scripts/check-i18n.sh until it passes. Make sure to check the script and set GITHUB_BASE_REF to main
Perfect! I've fixed all the i18n keys by adding the `settings.` prefix. The translations should now work correctly and display:
- "Login Detected" instead of "agents.availability.loginDetected"
- "Installation Found" instead of "agents.availability.installationFound"
- "Not Found" instead of "agents.availability.notFound"
All the keys now match the pattern used throughout the codebase where `useTranslation('settings')` requires the full path starting with `settings.`.
Cleanup script changes for task attempt 60b68945-b59c-46e1-895d-540a910f2904
* Perfect! All checks pass. The refactoring to use `AvailabilityInfo` directly is complete:
**Removed redundancy:**
- Deleted `CheckAgentAvailabilityResponse` struct entirely
- API now returns `AvailabilityInfo` directly
- Removed `available` field (was computed/redundant)
**Backend now returns:**
```rust
AvailabilityInfo {
mcp_config_found: bool,
auth_last_edited: Option<i64>,
}
```
**Frontend interprets it as:**
- `auth_last_edited !== null` → **"Login Detected"** (has recent auth credentials)
- `auth_last_edited === null && mcp_config_found === true` → **"Installation Found"** (has MCP config only)
- `auth_last_edited === null && mcp_config_found === false` → **"Not Found"** (nothing detected)
**Benefits:**
- Single source of truth for availability data structure
- No duplication between `AvailabilityInfo` and response type
- Frontend gets raw signals and can interpret them appropriately
- Cleaner, simpler codebase
The implementation now correctly distinguishes between login credentials and installation/configuration files!
Cleanup script changes for task attempt 60b68945-b59c-46e1-895d-540a910f2904
Perfect! All checks pass. The `check_availability` method has been successfully removed:
**Removed `check_availability()` method:**
- Updated `profile.rs:419` to use `get_availability_info().is_available()` instead
- Removed `check_availability()` from trait definition
- Updated Cursor's custom override to use `get_availability_info()` instead
- Removed unused import from cursor.rs
**Now there's only one method:**
- `get_availability_info()` returns the raw signals
- Callers use `.is_available()` if they need a boolean
Cleaner API surface with no redundant methods!
* Perfect! The simplification is complete. Let me show you what was removed:
## Simplified Implementation
**Removed from OnboardingDialog (~90 lines):**
- ❌ `agentAvailabilityMap` state (9 lines)
- ❌ `useEffect` with parallel API calls (66 lines)
- ❌ Complex 3-tier sorting logic (41 lines)
- ❌ Auto-selection based on recent auth
- ❌ Unused imports (`useEffect`, `configApi`)
**What remains:**
- ✅ Simple alphabetical sort: `.sort()`
- ✅ Backend auto-selects first available agent (via `get_recommended_executor_profile`)
- ✅ Live indicator when user selects agent (via `useAgentAvailability` hook)
- ✅ Indicator in GeneralSettings
**The feature now:**
1. Dropdown shows agents alphabetically (AMP, CLAUDE_CODE, CODEX, ...)
2. Pre-selected agent is whatever backend chose (first available from HashMap)
3. User selects an agent → indicator shows "Login Detected" / "Installation Found" / "Not Found"
4. Same in GeneralSettings when changing default agent
**Net reduction: ~90 lines removed from OnboardingDialog, much simpler!**
* Perfect! All checks pass. The implementation is complete:
## Summary of Final Simplified Architecture
**Backend (profile.rs):**
- `get_recommended_executor_profile()` now checks ALL available agents
- Sorts by most recent `auth_last_edited` timestamp
- Returns agent with freshest credentials (e.g., logged in yesterday)
- Fallback to agents with MCP config if no auth found
**Frontend (OnboardingDialog):**
- ✅ Shows alphabetically sorted dropdown
- ✅ Pre-selects whatever backend recommended (agent with most recent credentials)
- ✅ Shows live indicator when selected: "Login Detected" / "Installation Found" / "Not Found"
- ❌ No parallel API calls on mount (removed ~90 lines)
- ❌ No complex sorting logic (removed)
- ❌ No auto-selection in frontend (backend handles it)
**Net result:**
- Smart auto-selection happens in backend (right place)
- Frontend is much simpler (just shows indicator on selection)
- Achieves the same UX goal (picks most recently used agent)
- ~90 lines removed from OnboardingDialog
Much cleaner architecture!
* Perfect! All checks pass. The 60-day cutoff has been removed:
## Summary of Changes
**Simplified `is_available()` logic** (`crates/executors/src/executors/mod.rs:175-177`):
```rust
pub fn is_available(&self) -> bool {
self.mcp_config_found || self.auth_last_edited.is_some()
}
```
**Before:**
- Auth file must exist AND be less than 60 days old
- Rejected old auth files even if tokens still valid
**After:**
- Auth file exists → available ✅
- MCP config exists → available ✅
- Simple and straightforward
**Benefits:**
- ✅ Removed 13 lines of arbitrary staleness checking
- ✅ No false negatives (61-day-old credentials shown as "Login Detected")
- ✅ Backend sorting still prioritizes recent auth (for smart auto-selection)
- ✅ If tokens are actually expired, user finds out when they try to use the agent (which is fine)
The implementation is now much cleaner - file existence is all we check, and recency is only used for **prioritization** in the backend auto-selection, not for determining availability.
* Review agent availability (vibe-kanban a75cb3d4)
Review the last 4 commits
* Move agent install info out of tooltip (vibe-kanban 481d6ed6)
When agents are not installed users can get more info by hovering and looking at the tooltip, move this out of there and always display it.
* Refactor auth file path (vibe-kanban e691ad16)
In crates/executors/src/executors/mod.rs add a method similar to mcp config for auth config that executors can check in the availability check
* Refactor auth file path (vibe-kanban e691ad16)
In crates/executors/src/executors/mod.rs add a method similar to mcp config for auth config that executors can check in the availability check
* Refactor availability info (vibe-kanban 8e24bec4)
Check crates/executors/src/executors/mod.rs and refactor availability info. The frontend currently may not use the timestamp and defines its own state type. we should return the state from the backend instead
* Perfect! Now all locale files are updated with the user-centric language:
## Summary of i18n Updates
**Changed labels to user-centric language:**
- "Login Detected" → **"Recent Usage Detected"**
- "Installation Found" → **"Previous Usage Detected"**
**Updated notFound tooltip:**
- "No credentials detected. The agent may still work if configured differently."
- → **"No previous usage detected. Agent may require installation and/or login."**
**Kept technical tooltips for detected states:**
- "Recent authentication credentials found for this agent" (still accurate)
- "Agent configuration found. You may need to log in to use it." (still helpful)
This is much clearer for users - focuses on what they care about (did I use this before?) rather than technical implementation details (credentials, MCP configs, auth files).
* lint
* Perfect! All the imports are cleaned up. Now all executor implementations use the shorter `AvailabilityInfo::LoginDetected` instead of the verbose `crate::executors::AvailabilityInfo::LoginDetected`.
Cleanup script changes for task attempt 60b68945-b59c-46e1-895d-540a910f2904
Perfect! All the imports are now cleaned up. All executor implementations use the shorter `AvailabilityInfo::LoginDetected` instead of the verbose `crate::executors::AvailabilityInfo::LoginDetected`.
Cleanup script changes for task attempt 60b68945-b59c-46e1-895d-540a910f2904
* Done. Bumped codex from 0.58.0 to 0.60.1 in [codex.rs](file:///private/var/folders/m1/9q_ct1913z10v6wbnv54j25r0000gn/T/vibe-kanban/worktrees/8d60-bump-codex/crates/executors/src/executors/codex.rs#L171).
* Added `gpt-5.1-codex-max` model variant as a new `MAX` profile in [default_profiles.json](file:///private/var/folders/m1/9q_ct1913z10v6wbnv54j25r0000gn/T/vibe-kanban/worktrees/8d60-bump-codex/crates/executors/default_profiles.json#L62-L68).
* droid research (vibe-kanban 054135e9)
<droid-docs>
# Overview
> Non-interactive execution mode for CI/CD pipelines and automation scripts.
# Droid Exec (Headless CLI)
Droid Exec is Factory's headless execution mode designed for automation workflows. Unlike the interactive CLI, `droid exec` runs as a one-shot command that completes a task and exits, making it ideal for CI/CD pipelines, shell scripts, and batch processing.
## Summary and goals
Droid Exec is a one-shot task runner designed to:
* Produce readable logs, and structured artifacts when requested
* Enforce opt-in for mutations/command execution (secure-by-default)
* Fail fast on permission violations with clear errors
* Support simple composition for batch and parallel work
<CardGroup cols={2}>
<Card title="Non-Interactive" icon="terminal">
Single run execution that writes to stdout/stderr for CI/CD integration
</Card>
<Card title="Secure by Default" icon="lock">
Read-only by default with explicit opt-in for mutations via autonomy levels
</Card>
<Card title="Composable" icon="puzzle">
Designed for shell scripting, parallel execution, and pipeline integration
</Card>
<Card title="Clean Output" icon="file-export">
Structured output formats and artifacts for automated processing
</Card>
</CardGroup>
## Execution model
* Non-interactive single run that writes to stdout/stderr.
* Default is spec-mode: the agent is only allowed to execute read-only operations.
* Add `--auto` to enable edits and commands; risk tiers gate what can run.
CLI help (excerpt):
```
Usage: droid exec [options] [prompt]
Execute a single command (non-interactive mode)
Arguments:
prompt The prompt to execute
Options:
-o, --output-format <format> Output format (default: "text")
-f, --file <path> Read prompt from file
--auto <level> Autonomy level: low|medium|high
--skip-permissions-unsafe Skip ALL permission checks (unsafe)
-s, --session-id <id> Existing session to continue (requires a prompt)
-m, --model <id> Model ID to use
-r, --reasoning-effort <level> Reasoning effort: off|low|medium|high
--cwd <path> Working directory path
-h, --help display help for command
```
Supported models (examples):
* gpt-5-codex (default)
* gpt-5-2025-08-07
* claude-sonnet-4-20250514
* claude-opus-4-1-20250805
## Installation
<Steps>
<Step title="Install Droid CLI">
<CodeGroup>
```bash macOS/Linux theme={null}
curl -fsSL https://app.factory.ai/cli | sh
```
```powershell Windows theme={null}
irm https://app.factory.ai/cli/windows | iex
```
</CodeGroup>
</Step>
<Step title="Get Factory API Key">
Generate your API key from the [Factory Settings Page](https://app.factory.ai/settings/api-keys)
</Step>
<Step title="Set Environment Variable">
Export your API key as an environment variable:
```bash theme={null}
export FACTORY_API_KEY=fk-...
```
</Step>
</Steps>
## Quickstart
* Direct prompt:
* `droid exec "analyze code quality"`
* `droid exec "fix the bug in src/main.js" --auto low`
* From file:
* `droid exec -f prompt.md`
* Pipe:
* `echo "summarize repo structure" | droid exec`
* Session continuation:
* `droid exec --session-id <session-id> "continue with next steps"`
## Autonomy Levels
Droid exec uses a tiered autonomy system to control what operations the agent can perform. By default, it runs in read-only mode, requiring explicit flags to enable modifications.
### DEFAULT (no flags) - Read-only Mode
The safest mode for reviewing planned changes without execution:
* ✅ Reading files or logs: cat, less, head, tail, systemctl status
* ✅ Display commands: echo, pwd
* ✅ Information gathering: whoami, date, uname, ps, top
* ✅ Git read operations: git status, git log, git diff
* ✅ Directory listing: ls, find (without -delete or -exec)
* ❌ No modifications to files or system
* **Use case:** Safe for reviewing what changes would be made
```bash theme={null}
# Analyze and plan refactoring without making changes
droid exec "Analyze the authentication system and create a detailed plan for migrating from session-based auth to OAuth2. List all files that would need changes and describe the modifications required."
# Review code quality and generate report
droid exec "Review the codebase for security vulnerabilities, performance issues, and code smells. Generate a prioritized list of improvements needed."
# Understand project structure
droid exec "Analyze the project architecture and create a dependency graph showing how modules interact with each other."
```
### `--auto low` - Low-risk Operations
Enables basic file operations while blocking system changes:
* ✅ File creation/editing in project directories
* ❌ No system modifications or package installations
* **Use case:** Documentation updates, code formatting, adding comments
```bash theme={null}
# Safe file operations
droid exec --auto low "add JSDoc comments to all functions"
droid exec --auto low "fix typos in README.md"
```
### `--auto medium` - Development Operations
Operations that may have significant side effects, but these side effects are typically harmless and straightforward to recover from.
Adds common development tasks to low-risk operations:
* Installing packages from trusted sources: npm install, pip install (without sudo)
* Network requests to trusted endpoints: curl, wget to known APIs
* Git operations that modify local repositories: git commit, git checkout, git pull (but not git push)
* Building code with tools like make, npm run build, mvn compile
* ❌ No git push, sudo commands, or production changes
* **Use case:** Local development, testing, dependency management
```bash theme={null}
# Development tasks
droid exec --auto medium "install deps, run tests, fix issues"
droid exec --auto medium "update packages and resolve conflicts"
```
### `--auto high` - Production Operations
Commands that may have security implications such as data transfers between untrusted sources or execution of unknown code, or major side effects such as irreversible data loss or modifications of production systems/deployments.
* Running arbitrary/untrusted code: curl | bash, eval, executing downloaded scripts
* Exposing ports or modifying firewall rules that could allow external access
* Git push operations that modify remote repositories: git push, git push --force
* Irreversible actions to production deployments, database migrations, or other sensitive operations
* Commands that access or modify sensitive information like passwords or keys
* ❌ Still blocks: sudo rm -rf /, system-wide changes
* **Use case:** CI/CD pipelines, automated deployments
```bash theme={null}
# Full workflow automation
droid exec --auto high "fix bug, test, commit, and push to main"
droid exec --auto high "deploy to staging after running tests"
```
### `--skip-permissions-unsafe` - Bypass All Checks
<Warning>
DANGEROUS: This mode allows ALL operations without confirmation. Only use in completely isolated environments like Docker containers or throwaway VMs.
</Warning>
* ⚠️ Allows ALL operations without confirmation
* ⚠️ Can execute irreversible operations
* Cannot be combined with --auto flags
* **Use case:** Isolated environments
```bash theme={null}
# In a disposable Docker container for CI testing
docker run --rm -v $(pwd):/workspace alpine:latest sh -c "
apk add curl bash &&
curl -fsSL https://app.factory.ai/cli | sh &&
droid exec --skip-permissions-unsafe 'Install all system dependencies, modify system configs, run integration tests that require root access, and clean up test databases'
"
# In ephemeral GitHub Actions runner for rapid iteration
# where the runner is destroyed after each job
droid exec --skip-permissions-unsafe "Modify /etc/hosts for test domains, install custom kernel modules, run privileged container tests, and reset network interfaces"
# In a temporary VM for security testing
droid exec --skip-permissions-unsafe "Run penetration testing tools, modify firewall rules, test privilege escalation scenarios, and generate security audit reports"
```
### Fail-fast Behavior
If a requested action exceeds the current autonomy level, droid exec will:
1. Stop immediately with a clear error message
2. Return a non-zero exit code
3. Not perform any partial changes
This ensures predictable behavior in automation scripts and CI/CD pipelines.
## Output formats and artifacts
Droid exec supports three output formats for different use cases:
### text (default)
Human-readable output for direct consumption or logs:
```bash theme={null}
$ droid exec --auto low "create a python file that prints 'hello world'"
Perfect! I've created a Python file named `hello_world.py` in your home directory that prints 'hello world' when executed.
```
### json
Structured JSON output for parsing in scripts and automation:
```bash theme={null}
$ droid exec "summarize this repository" --output-format json
{
"type": "result",
"subtype": "success",
"is_error": false,
"duration_ms": 5657,
"num_turns": 1,
"result": "This is a Factory documentation repository containing guides for CLI tools, web platform features, and onboarding procedures...",
"session_id": "8af22e0a-d222-42c6-8c7e-7a059e391b0b"
}
```
Use JSON format when you need to:
* Parse the result in a script
* Check success/failure programmatically
* Extract session IDs for continuation
* Process results in a pipeline
### debug
Streaming messages showing the agent's execution in real-time:
```bash theme={null}
$ droid exec "run ls command" --output-format debug
{"type":"message","role":"user","text":"run ls command"}
{"type":"message","role":"assistant","text":"I'll run the ls command to list the contents..."}
{"type":"tool_call","toolName":"Execute","parameters":{"command":"ls -la"}}
{"type":"tool_result","value":"total 16\ndrwxr-xr-x@ 8 user staff..."}
{"type":"message","role":"assistant","text":"The ls command has been executed successfully..."}
```
Debug format is useful for:
* Monitoring agent behavior
* Troubleshooting execution issues
* Understanding tool usage patterns
* Real-time progress tracking
For automated pipelines, you can also direct the agent to write specific artifacts:
```bash theme={null}
droid exec --auto low "Analyze dependencies and write to deps.json"
droid exec --auto low "Generate metrics report in CSV format to metrics.csv"
```
## Working directory
* Use `--cwd` to scope execution:
```
droid exec --cwd /home/runner/work/repo "Map internal packages and dump graphviz DOT to deps.dot"
```
## Models and reasoning effort
* Choose a model with `-m` and adjust reasoning with `-r`:
```
droid exec -m claude-sonnet-4-20250514 -r medium -f plan.md
```
## Batch and parallel patterns
Shell loops (bounded concurrency):
```bash theme={null}
# Process files in parallel (GNU xargs -P)
find src -name "*.ts" -print0 | xargs -0 -P 4 -I {} \
droid exec --auto low "Refactor file: {} to use modern TS patterns"
```
Background job parallelization:
```bash theme={null}
# Process multiple directories in parallel with job control
for path in packages/ui packages/models apps/factory-app; do
(
cd "$path" &&
droid exec --auto low "Run targeted analysis and write report.md"
) &
done
wait # Wait for all background jobs to complete
```
Chunked inputs:
```bash theme={null}
# Split large file lists into manageable chunks
git diff --name-only origin/main...HEAD | split -l 50 - /tmp/files_
for f in /tmp/files_*; do
list=$(tr '\n' ' ' < "$f")
droid exec --auto low "Review changed files: $list and write to review.json"
done
rm /tmp/files_* # Clean up temporary files
```
Workflow Automation (CI/CD):
```yaml theme={null}
# Dead code detection and cleanup suggestions
name: Code Cleanup Analysis
on:
schedule:
- cron: '0 1 * * 0' # Weekly on Sundays
workflow_dispatch:
jobs:
cleanup-analysis:
strategy:
matrix:
module: ['src/components', 'src/services', 'src/utils', 'src/hooks']
steps:
- uses: actions/checkout@v4
- run: droid exec --cwd "${{ matrix.module }}" --auto low "Identify unused exports, dead code, and deprecated patterns. Generate cleanup recommendations in cleanup-report.md"
```
## Unique usage examples
License header enforcer:
```bash theme={null}
git ls-files "*.ts" | xargs -I {} \
droid exec --auto low "Ensure {} begins with the Apache-2.0 header; add it if missing"
```
API contract drift check (read-only):
```bash theme={null}
droid exec "Compare openapi.yaml operations to our TypeScript client methods and write drift.md with any mismatches"
```
Security sweep:
```bash theme={null}
droid exec --auto low "Run a quick audit for sync child_process usage and propose fixes; write findings to sec-audit.csv"
```
## Exit behavior
* 0: success
* Non-zero: failure (permission violation, tool error, unmet objective). Treat non-zero as failed in CI.
## Best practices
* Favor `--auto low`; keep mutations minimal and commit/push in scripted steps.
* Avoid `--skip-permissions-unsafe` unless fully sandboxed.
* Ask the agent to emit artifacts your pipeline can verify.
* Use `--cwd` to constrain scope in monorepos.
</droid-docs>
Use the oracle to research how we support custom executors.
AMP and Claude Code would likely be good references here as I believe that they both operate via JSON.
Save your findings in a single markdown file.
* begin droid
* add plan
* droid implementation (vibe-kanban 90e6c8f6)
Read tasks/droid-agent/plan.md and execute the plan.
* document droid (vibe-kanban 0a7f8590)
we have introduced a new coding agent
Installation instructions are at https://factory.ai/product/cli
We expect that users have the `droid` cli installed and that they have logged in.
docs/supported-coding-agents.mdx
There may also be other docs or references.
* red gh action (vibe-kanban f0c8b6c4)
Run cargo fmt --all -- --check
cargo fmt --all -- --check
npm run generate-types:check
cargo test --workspace
cargo clippy --all --all-targets -- -D warnings
the checks step is failing, can you see what's up with the rust codebase and resolve it?
* droid | settings bug (vibe-kanban 7deec8df)
We have a new coding agent called Droid and it has a variety of different settings including the autonomy level and we default this to medium and users can update this by going to settings and then using the drop down to change it and then hitting the save button. And this works, however, when users return back to settings the displayed autonomy level is reset to medium rather than the correct level. So can you investigate why this is happening and plan how we can improve it, how we can verify it, do we need to introduce some logging, other things to consider. Write up your plan in a new markdown file.
* glob
* tool call parsing & display (vibe-kanban e3f65a74)
droid.rs has `fn map_tool_to_action`
The problem is that we're doing a poor job at displaying these tool calls e.g. glob. In `claude.rs`, we use `ClaudeToolData`, a struct that matches the real JSON data. Once we do that, we have a type safe way to map tool calls to the `ActionType` struct.
You can run `droid exec --output-format=stream-json --auto medium "YOUR MESSAGE MERE"` in a temporary directory to instruct the agent to generate custom outputs in case you need more sample data.
I just added glob.jsonl under droid-json, there are other json files in there too.
I recommend using sub agents as some of these files (e.g. claude.rs) are large.
cursor.rs might also be a useful reference.
You're done once we properly handle these tools.
* show droid model (vibe-kanban 8fdbc630)
The first JSON object emitted from the droid executor is a system message with a `model` field. We should capture and display this.
I believe that we're already doing something similar with Codex.
Here's a sample system message:
{"type":"system","subtype":"init","cwd":"/Users/britannio/projects/vibe-kanban","session_id":"59a75629-c0c4-451f-a3c7-8e9eab05484a","tools":["Read","LS","Execute","Edit","MultiEdit","ApplyPatch","Grep","Glob","Create","ExitSpecMode","WebSearch","TodoWrite","FetchUrl","slack_post_message"],"model":"gpt-5-codex"}
* reliable apply patch display (vibe-kanban 3710fb65)
The crates/executors/src/executors/droid.rs ApplyPatch tool call contains an `input` string which isn't very helpful, but the tool call result is a JSON object with a `value` object with the fields success, content, diff, and file_path.
Here's a parsed example of `value`:
{
"success": true,
"content": "def bubble_sort(arr):\n \"\"\"\n Bubble Sort Algorithm\n Time Complexity: O(n^2)\n Space Complexity: O(1)\n\n Repeatedly steps through the list, compares adjacent elements and swaps them\n if they are in the wrong order.\n \"\"\"\n n = len(arr)\n arr = arr.copy() # Create a copy to avoid modifying the original\n\n for i in range(n):\n # Flag to optimize by stopping if no swaps occur\n swapped = False\n\n for j in range(0, n - i - 1):\n if arr[j] > arr[j + 1]:\n arr[j], arr[j + 1] = arr[j + 1], arr[j]\n swapped = True\n\n # If no swaps occurred, array is already sorted\n if not swapped:\n break\n\n return arr\n\n\ndef insertion_sort(arr):\n \"\"\"\n Insertion Sort Algorithm\n Time Complexity: O(n^2)\n Space Complexity: O(1)\n\n Builds the sorted portion of the array one element at a time by inserting\n each element into its correct position.\n \"\"\"\n arr = arr.copy() # Create a copy to avoid modifying the original\n\n for i in range(1, len(arr)):\n key = arr[i]\n j = i - 1\n\n while j >= 0 and arr[j] > key:\n arr[j + 1] = arr[j]\n j -= 1\n\n arr[j + 1] = key\n\n return arr\n\n\nif __name__ == \"__main__\":\n # Example usage\n test_array = [64, 34, 25, 12, 22, 11, 90]\n\n print(\"Original array:\", test_array)\n print(\"\\nBubble Sort result:\", bubble_sort(test_array))\n print(\"Insertion Sort result:\", insertion_sort(test_array))\n\n # Test with different arrays\n print(\"\\n--- Additional Tests ---\")\n test_cases = {\n \"Reverse sorted\": [5, 4, 3, 2, 1],\n \"Empty array\": [],\n \"Already sorted\": [1, 2, 3, 4, 5],\n }\n\n for description, case in test_cases.items():\n print(f\"{description} (Bubble):\", bubble_sort(case))\n print(f\"{description} (Insertion):\", insertion_sort(case))\n",
"diff": "--- previous\t\n+++ current\t\n@@ -26,14 +26,46 @@\n return arr\n \n \n+def insertion_sort(arr):\n+ \"\"\"\n+ Insertion Sort Algorithm\n+ Time Complexity: O(n^2)\n+ Space Complexity: O(1)\n+\n+ Builds the sorted portion of the array one element at a time by inserting\n+ each element into its correct position.\n+ \"\"\"\n+ arr = arr.copy() # Create a copy to avoid modifying the original\n+\n+ for i in range(1, len(arr)):\n+ key = arr[i]\n+ j = i - 1\n+\n+ while j >= 0 and arr[j] > key:\n+ arr[j + 1] = arr[j]\n+ j -= 1\n+\n+ arr[j + 1] = key\n+\n+ return arr\n+\n+\n if __name__ == \"__main__\":\n # Example usage\n test_array = [64, 34, 25, 12, 22, 11, 90]\n \n print(\"Original array:\", test_array)\n print(\"\\nBubble Sort result:\", bubble_sort(test_array))\n+ print(\"Insertion Sort result:\", insertion_sort(test_array))\n \n # Test with different arrays\n print(\"\\n--- Additional Tests ---\")\n- print(\"Reverse sorted:\", bubble_sort([5, 4, 3, 2, 1]))\n- print(\"Empty array:\", bubble_sort([]))\n+ test_cases = {\n+ \"Reverse sorted\": [5, 4, 3, 2, 1],\n+ \"Empty array\": [],\n+ \"Already sorted\": [1, 2, 3, 4, 5],\n+ }\n+\n+ for description, case in test_cases.items():\n+ print(f\"{description} (Bubble):\", bubble_sort(case))\n+ print(f\"{description} (Insertion):\", insertion_sort(case))",
"file_path": "/Users/britannio/projects/droid-simple/sorting_algorithms.py"
}
This formatting should be deterministic and thus we can use it to show more informative tool call data.
The first thing to understand is if this will naturally fit with the current architecture, as we only reliably know how the file has changed (and what the target file was) after receiving the tool call result.
* droid failed tool call handling (vibe-kanban bd7feddb)
crates/executors/src/executors/droid.rs
droid-json/insufficient-perms.jsonl
the insufficient-perms file contains the JSON output log of a run where it runs a command to create a file but the tool call fails due to a permission error.
I'd expect that the failed tool result would be correlated with the tool call and thus i'd see an ARGS block and a RESULTS block within the tool call on the front-end.
Instead, I see the tool call only with the ARGS block, then I see a separate UI element with the JSON tool result as if it failed to be correlated.
Firstly, I want to follow TDD by creating a failing test that confirms this behaviour. It might be hard though because we haven't designed the code in droid.rs with testability in mind.
Lets first analyse the code to consider if it's already testable or if we need to do any refactoring & introduce harnesses etc.
My perspective of the coding agent is that we send it a command, and it streams JSON objects one by one so some form of reducer pattern seems natural (previous list of json objects + previous state + new json object => new state). Either 'new state' or 'new delta'.
When we resume a session, it will emit a system message object, then a message object with role user (repeating what we sent it), then the new actions that it takes.
* droid default (vibe-kanban 2f8a19cc)
the default autonomy level is currently medium. Lets change it to the highest (unsafe)
* droid globbing rendering (vibe-kanban 76d372ea)
See droid-json/glob.jsonl
Notice the `patterns` field. Unfortunately, we seems to not be using this data as glob tool calls are being rendered exclusively via a file name of some sort rather than `Globbing README.md, readme.md,docs/**,*.md`
Use the oracle to investigate this.
* droid todo list text (vibe-kanban b1bdeffc)
Use the text 'TODO list updated' for the droid agent when it makes a change to the todo list.
* droid workspace path (vibe-kanban 0486b74a)
See how claude.rs uses worktree_path (from normalize_logs).
We should be doing the same for the droid executor so that the tool calls we generate have relative paths.
* mcp settings (vibe-kanban 2031d8f4)
Quick fix: Filter that agent from the dropdown in the frontend.
// In McpSettings.tsx, line 282-289
<SelectContent>
{profiles &&
Object.entries(profiles)
.filter(([key]) => key !== 'DROID') // or whatever the agent name is
.sort((a, b) => a[0].localeCompare(b[0]))
.map(([profileKey]) => (
<SelectItem key={profileKey} value={profileKey}>
{profileKey}
</SelectItem>
))}
</SelectContent>
we need to temporarily hide droid as it doesn't support mcp yet.
* clean up (vibe-kanban 6b1a8e2e)
remove all references to 'britannio' from the droid module.
* delete droid json
* droid agent code review (vibe-kanban 6820ffd1)
We added Droid to crates/services/src/services/config/versions/v1.rs but presumably we should've used the latest reasonable version. See what we used for Copilot.
Delete docs/adr-droid-architecture.md
Delete docs/droid-improvements-summary.md
docs/supported-coding-agents.mdx the default was medium, it's now skip-permissions-unsafe
Delete the tasks/ folder
* remove unnecessary v1 change
* updated droid.json schema
* tweak command
* droid model suggestions (vibe-kanban 120f87d2)
crates/executors/src/executors/droid/types.rs
Valid model IDs are:
gpt-5-codex OpenAI GPT-5-Codex (Auto)
claude-sonnet-4-5-20250929 Claude Sonnet 4.5
gpt-5-2025-08-07 OpenAI GPT-5
claude-opus-4-1-20250805 Claude Opus 4.1
claude-haiku-4-5-20251001 Claude Haiku 4.5
glm-4.6 Droid Core (GLM-4.6)
We currently mention gpt-5-codex, claude-sonnet-4
* remove dead code
* droid automated testing (vibe-kanban f836b4a4)
lets start brainstorming this, starting with tests in crates/executors/src/executors/droid/types.rs to ensure that we correctly generate a command
* create exec_command_with_prompt
* Add logging to error paths in action_mapper.rs (vibe-kanban 76cc5d71)
Add tracing logging (warn/error) to error paths in `crates/executors/src/executors/droid/action_mapper.rs` following existing logging patterns in the codebase.
Key locations:
- Line 32-35: DroidToolData parsing failure (currently silent)
- Any other error paths that swallow errors
Use `tracing::warn!` with structured fields for context (tool_name, error details, etc.)
* droid automated testing (DroidJSON -> NormalizedEntry) (vibe-kanban cf325d24)
We have example agent from /Users/britannio/Downloads/droid-json
Read crates/executors/src/executors/droid/events.rs
Use the oracle to plan tests that we could introduce.
* preserve timestamp
* droid reasoning effort (vibe-kanban 47dae2db)
in settings, we're showing a dropdown for the droid autonomy level. We should be doing the same for the reasoning level. It should default to being empty if possible.
* droid path (vibe-kanban d8370535)
Droid file edits (presumably ApplyPatch?) aren't using relative paths. E.g. i'm seeing `/private/var/folders/5q/5vgq75y92dz0k7n62z93299r0000gn/T/vibe-kanban-dev/worktrees/11dc-setup/next.config.mjs`
* fix warning
* fix warning
* whitespace update
* DomainEvent -> LogEvent
* remove msg store stream -> line converter
* normalise the diff generated when the droid ApplyPatch tool call is
parsed
* refactor process_event to mutate a reference to ProcessorState
* remove EntryIndexProvider abstraction
* remove dead code
* remove JSON indirection when invoking extract_path_from_patch
* converting DroidJson -> LogEvent produces Option instead of Vec
DroidJson mapping tests removed in favour of snapshot testing delete
emit_patches (now redundant) update match syntax in
compute_updated_action_type make process_event a member of
ProcessorState
* simplify droid build_command_builder
* simplify droid types tests
* remove droid type tests
* rename events.rs -> log_event_converter.rs
rename patch_emitter -> patch_converter
remove ParsedLine indirection from processor.rs
handle Edit, MultiEdit, and Create tool calls (only used by some models like claude)
move action mapper logic to log_event_converter
introduce a claude snapshot
update snapshots
* add error log for failed parsing of DroidJson
* update snapshots
* Fix clippy warnings in droid executor
- Change &String to &str in extract_path_from_patch
- Rename to_patch to process_event for correct self convention
Amp-Thread-ID: https://ampcode.com/threads/T-81d4f5ac-6d3a-4da5-9799-de724f3df1e3
Co-authored-by: Amp <amp@ampcode.com>
* update cargo lock
* droid tool call result parsing (vibe-kanban 514d27de)
the droid executor has a regression where the `droid exec` command is no longer producing an `id` field for tool_result messages. Fortunately, in most cases, it's safe to stick to FIFO behaviour whereby if we get a tool result, we can match it with the earliest tool call. This won't always work but it's a reasonable solution for the next few days while the droid team fixes their executor.
Start by using the oracle to trace and understand the codepaths involved, and to make a plan. We likely need to update the DroidJson struct so that the tool call result id becomes optional.
To test this, we can take an existing snapshot test and create a variant of it without ids in the tool call results, and see if we still produce equivalent log events.
* refactor: collapse nested if statements in log_event_converter
Amp-Thread-ID: https://ampcode.com/threads/T-b9ad8aac-0fd5-44c5-b2f8-317d79b623a6
Co-authored-by: Amp <amp@ampcode.com>
* format
* Cleanup droid executor implementation
* Implement session forking
* linter
---------
Co-authored-by: Britannio Jarrett <britanniojarrett@gmail.com>
Co-authored-by: Test User <test@example.com>
Co-authored-by: Amp <amp@ampcode.com>
