fix: Keep V3 CLI JSON output compatible with V2

.agents/skills/uloop-clear-console/SKILL.md

@@ -39,10 +39,11 @@ uloop clear-console --add-confirmation-message
 ## Output
 
 Returns JSON with:
-- `success` (boolean): Whether the clear operation succeeded
-- `clearedLogCount` (number): Total number of log entries that were cleared
-- `clearedCounts` (object): Breakdown by log type
-  - `errorCount` (number): Errors cleared
-  - `warningCount` (number): Warnings cleared
-  - `logCount` (number): Info logs cleared
-- `message` (string): Description of the result. On failure, this field carries the error summary.
+
+- `Success` (boolean): Whether the clear operation succeeded
+- `ClearedLogCount` (number): Total number of log entries that were cleared
+- `ClearedCounts` (object): Breakdown by log type
+  - `ErrorCount` (number): Errors cleared
+  - `WarningCount` (number): Warnings cleared
+  - `LogCount` (number): Info logs cleared
+- `Message` (string): Description of the result. On failure, this field carries the error summary.

.agents/skills/uloop-compile/SKILL.md

@@ -44,7 +44,8 @@ uloop compile --stop-on-external-scene-changes
 ## Output
 
 Returns JSON:
-- `success`: boolean or null
-- `errorCount`: number or null
-- `warningCount`: number or null
-- `message`: string
+
+- `Success`: boolean or null
+- `ErrorCount`: number or null
+- `WarningCount`: number or null
+- `Message`: string

.agents/skills/uloop-control-play-mode/SKILL.md

@@ -49,14 +49,15 @@ uloop control-play-mode --action Step
 ## Output
 
 Returns JSON with the current play mode state:
-- `isPlaying`: Whether Unity is currently in play mode
-- `isPaused`: Whether play mode is paused
-- `changed`: Whether the requested action changed the current play mode state
-- `wasAlreadyStopped`: Whether `Stop` was requested while Play Mode was already stopped
-- `message`: Description of the action performed
+
+- `IsPlaying`: Whether Unity is currently in play mode
+- `IsPaused`: Whether play mode is paused
+- `Changed`: Whether the requested action changed the current play mode state
+- `WasAlreadyStopped`: Whether `Stop` was requested while Play Mode was already stopped
+- `Message`: Description of the action performed
 
 ## Notes
 
-- Stop on an already-stopped Editor sets `changed: false`, `wasAlreadyStopped: true`
+- Stop on an already-stopped Editor sets `Changed: false`, `WasAlreadyStopped: true`
 - Step is independent of `Time.timeScale`; requires PlayMode to be running
 - The command waits for the requested state before returning. Increase `--timeout-seconds` for projects with slow PlayMode entry.

.agents/skills/uloop-execute-dynamic-code/SKILL.md

@@ -43,16 +43,17 @@ Prefer terminal commands for file operations and keep snippets focused on Unity
 ## Output
 
 Returns JSON:
-- `success`: boolean — overall execution success
-- `result`: string — value of the snippet's `return` statement (empty when omitted)
-- `logs`: string[] — execution messages from the dynamic-code tool; read Unity Console `Debug.Log` output with `get-logs`
-- `compilationErrors`: object[] — Roslyn diagnostics with `message`, `line`, `column`, `errorCode`, optional `hint` and `suggestions`
-- `error` / `errorMessage`: string — top-level failure summary (empty on success)
-- `updatedCode`: string|null — the wrapped form actually compiled (handy when debugging using-statement reordering)
-- `diagnosticsSummary`: string|null — compact summary when diagnostics are available
-- `diagnostics`: object[] — structured diagnostics; same shape as `compilationErrors`, usually populated together with it
-
-On `success: false`, inspect `compilationErrors` first. If empty, read `errorMessage` (and `logs` for extra context) — the failure may be a runtime exception, cancellation, or an "execution in progress" rejection, all of which return empty `compilationErrors`. Both EditMode and PlayMode are supported targets — the snippet runs in whichever mode the Editor is currently in.
+
+- `Success`: boolean — overall execution success
+- `Result`: string — value of the snippet's `return` statement (empty when omitted)
+- `Logs`: string[] — execution messages from the dynamic-code tool; read Unity Console `Debug.Log` output with `get-logs`
+- `CompilationErrors`: object[] — Roslyn diagnostics with `Message`, `Line`, `Column`, `ErrorCode`, optional `Hint` and `Suggestions`
+- `Error` / `ErrorMessage`: string — top-level failure summary (empty on success)
+- `UpdatedCode`: string|null — the wrapped form actually compiled (handy when debugging using-statement reordering)
+- `DiagnosticsSummary`: string|null — compact summary when diagnostics are available
+- `Diagnostics`: object[] — structured diagnostics; same shape as `CompilationErrors`, usually populated together with it
+
+On `Success: false`, inspect `CompilationErrors` first. If empty, read `ErrorMessage` (and `Logs` for extra context) — the failure may be a runtime exception, cancellation, or an "execution in progress" rejection, all of which return empty `CompilationErrors`. Both EditMode and PlayMode are supported targets — the snippet runs in whichever mode the Editor is currently in.
 
 ## Code Examples by Category
 

.agents/skills/uloop-find-game-objects/SKILL.md

@@ -70,21 +70,23 @@ uloop find-game-objects --search-mode Selected --include-inactive
 ## Output
 
 Returns JSON with:
-- `results` (array): Matching GameObjects, each containing:
-  - `name` (string): GameObject name
-  - `path` (string): Hierarchy path (e.g., `Canvas/Panel/Button`)
-  - `isActive` (boolean): Active state in hierarchy
-  - `tag` (string): GameObject tag
-  - `layer` (number): Layer index
-  - `components` (array): Each entry has `type` (short name, e.g., `Rigidbody`), `fullTypeName` (e.g., `UnityEngine.Rigidbody`), and `properties` (array of Inspector-visible `{name, type, value}` pairs)
-- `totalFound` (number): Results returned (after `--max-results` clipping). For multi-selection file export, this is the number exported.
-- `errorMessage` (string): Top-level failure summary (empty on success)
-- `processingErrors` (array): Selected-mode per-GameObject serialization failures, each `{gameObjectName, gameObjectPath, error}`. Omitted/null or empty on clean runs.
+
+- `Results` (array): Matching GameObjects, each containing:
+  - `Name` (string): GameObject name
+  - `Path` (string): Hierarchy path (e.g., `Canvas/Panel/Button`)
+  - `IsActive` (boolean): Active state in hierarchy
+  - `Tag` (string): GameObject tag
+  - `Layer` (number): Layer index
+  - `Components` (array): Each entry has `Type` (short name, e.g., `Rigidbody`), `FullTypeName` (e.g., `UnityEngine.Rigidbody`), and `Properties` (array of Inspector-visible `{Name, Type, Value}` pairs)
+- `TotalFound` (number): Results returned (after `--max-results` clipping). For multi-selection file export, this is the number exported.
+- `ErrorMessage` (string): Top-level failure summary (empty on success)
+- `ProcessingErrors` (array): Selected-mode per-GameObject serialization failures, each `{GameObjectName, GameObjectPath, Error}`. Omitted/null or empty on clean runs.
 
 ### Multi-selection file export
 
-For `Selected` mode with **multiple** successfully serialized GameObjects, inline `results` is not populated and the data is written to a file instead. Two extra fields appear:
-- `resultsFilePath` (string): Relative path under `.uloop/outputs/FindGameObjectsResults/`
-- `message` (string): Human-readable summary (e.g., "5 GameObjects exported")
+For `Selected` mode with **multiple** successfully serialized GameObjects, inline `Results` is not populated and the data is written to a file instead. Two extra fields appear:
+
+- `ResultsFilePath` (string): Relative path under `.uloop/outputs/FindGameObjectsResults/`
+- `Message` (string): Human-readable summary (e.g., "5 GameObjects exported")
 
-Single-selection and search-mode calls (`Exact`, `Path`, `Regex`, `Contains`) always return inline. No selection (`Selected` mode with empty selection) returns empty `results` plus a `message`.
+Single-selection and search-mode calls (`Exact`, `Path`, `Regex`, `Contains`) always return inline. No selection (`Selected` mode with empty selection) returns empty `Results` plus a `Message`.

.agents/skills/uloop-focus-window/SKILL.md

@@ -29,8 +29,9 @@ uloop focus-window
 ## Output
 
 Returns JSON with:
-- `success`: Whether the focus operation succeeded
-- `message`: Status message (e.g. `Unity Editor window focused (PID: 12345)`, or the failure reason such as `Unity project not found` / `No running Unity process found for this project` / `Failed to focus Unity window: <reason>`)
+
+- `Success`: Whether the focus operation succeeded
+- `Message`: Status message (e.g. `Unity Editor window focused (PID: 12345)`, or the failure reason such as `Unity project not found` / `No running Unity process found for this project` / `Failed to focus Unity window: <reason>`)
 
 ## Notes
 

.agents/skills/uloop-get-hierarchy/SKILL.md

@@ -56,7 +56,8 @@ uloop get-hierarchy --use-selection
 ## Output
 
 Returns JSON with:
-- `message` (string): Human-readable guidance pointing at the saved file
-- `hierarchyFilePath` (string): Filesystem path to the JSON file that contains the actual hierarchy data
 
-The hierarchy itself is **not** in the response — it is written to the file at `hierarchyFilePath`. Open that file to read the `Context` and `Hierarchy` payload (GameObject tree, components, etc.).
+- `Message` (string): Human-readable guidance pointing at the saved file
+- `HierarchyFilePath` (string): Filesystem path to the JSON file that contains the actual hierarchy data
+
+The hierarchy itself is **not** in the response — it is written to the file at `HierarchyFilePath`. Open that file to read the `Context` and `Hierarchy` payload (GameObject tree, components, etc.).

.agents/skills/uloop-get-logs/SKILL.md

@@ -50,10 +50,11 @@ uloop get-logs --search-text "Missing.*Component" --use-regex
 ## Output
 
 Returns JSON with:
-- `totalCount` (number): Total logs available before max-count clipping
-- `displayedCount` (number): Logs returned in this response (≤ `--max-count`)
-- Input filters (`logType`, `maxCount`, `searchText`, `includeStackTrace`) are echoed back in the response
-- `logs` (array): Each entry has:
-  - `type` (string): `"Error"`, `"Warning"`, or `"Log"`
-  - `message` (string): Log message body
-  - `stackTrace` (string): Stack trace text. Empty when `--include-stack-trace` is `false`.
+
+- `TotalCount` (number): Total logs available before max-count clipping
+- `DisplayedCount` (number): Logs returned in this response (≤ `--max-count`)
+- Input filters (`LogType`, `MaxCount`, `SearchText`, `IncludeStackTrace`) are echoed back in the response
+- `Logs` (array): Each entry has:
+  - `Type` (string): `"Error"`, `"Warning"`, or `"Log"`
+  - `Message` (string): Log message body
+  - `StackTrace` (string): Stack trace text. Empty when `--include-stack-trace` is `false`.

.agents/skills/uloop-hello-world/SKILL.md

@@ -40,13 +40,15 @@ uloop hello-world --name "Carlos" --language spanish --no-include-timestamp
 ## Output
 
 Returns JSON with:
-- `message`: The greeting message
-- `language`: Language used for greeting
-- `timestamp`: Current timestamp (if enabled)
+
+- `Message`: The greeting message
+- `Language`: Language used for greeting
+- `Timestamp`: Current timestamp (if enabled)
 
 ## Notes
 
 This is a sample custom tool demonstrating:
+
 - Type-safe parameter handling with Schema
 - Enum parameters for language selection
 - Value-less flag parameters

.agents/skills/uloop-launch/SKILL.md

@@ -50,18 +50,19 @@ uloop launch --quit
 May print status/progress lines before the final JSON payload, such as project path, detected Unity version, or readiness wait messages.
 
 The final JSON payload includes:
-- `success`: whether the command completed
-- `ready`: whether Unity CLI Loop is ready for commands
-- `serverReady`: whether the Unity CLI Loop server accepted requests
-- `projectIpcReady`: whether the project IPC path accepted tool requests
-- `alreadyRunning`: whether an existing Unity process was reused
-- `launched`: whether this command launched a Unity process
-- `restarted`: whether this command stopped an existing process and launched a new one
-- `quit`: whether this command stopped Unity without launching a new process
-- `previousProcessId`: process ID stopped by restart or quit, when available
-- `currentProcessId`: current Unity process ID, when available
-- `projectRoot`: resolved project root
-- `message`: readiness summary
+
+- `Success`: whether the command completed
+- `Ready`: whether Unity CLI Loop is ready for commands
+- `ServerReady`: whether the Unity CLI Loop server accepted requests
+- `ProjectIpcReady`: whether the project IPC path accepted tool requests
+- `AlreadyRunning`: whether an existing Unity process was reused
+- `Launched`: whether this command launched a Unity process
+- `Restarted`: whether this command stopped an existing process and launched a new one
+- `Quit`: whether this command stopped Unity without launching a new process
+- `PreviousProcessId`: process ID stopped by restart or quit, when available
+- `CurrentProcessId`: current Unity process ID, when available
+- `ProjectRoot`: resolved project root
+- `Message`: readiness summary
 
 ## Notes
 

.agents/skills/uloop-record-input/SKILL.md

@@ -44,11 +44,12 @@ Replay injects input frame-by-frame, so the game must also be deterministic to p
 ## Output
 
 The CLI prints JSON with:
-- `success`: Whether the operation succeeded
-- `message`: Status message
-- `action`: Echoes which action was executed (`Start` or `Stop`)
-- `outputPath`: Path to saved recording (nullable; populated on `Stop` only)
-- `totalFrames`: Number of frames recorded (nullable int; populated on `Stop` only)
-- `durationSeconds`: Recording duration in seconds (nullable float; populated on `Stop` only)
-
-Frame data lives only in the JSON file at `outputPath`; internal metadata fields are not included in the response.
+
+- `Success`: Whether the operation succeeded
+- `Message`: Status message
+- `Action`: Echoes which action was executed (`Start` or `Stop`)
+- `OutputPath`: Path to saved recording (nullable; populated on `Stop` only)
+- `TotalFrames`: Number of frames recorded (nullable int; populated on `Stop` only)
+- `DurationSeconds`: Recording duration in seconds (nullable float; populated on `Stop` only)
+
+Frame data lives only in the JSON file at `OutputPath`; internal metadata fields are not included in the response.

.agents/skills/uloop-replay-input/SKILL.md

@@ -48,11 +48,12 @@ Replay injects the exact same input frame-by-frame, but the game must also be de
 ## Output
 
 Returns JSON with:
-- `success`: Whether the operation succeeded
-- `message`: Status message
-- `action`: Echoes which action was executed (`Start`, `Stop`, or `Status`)
-- `inputPath`: Path to recording file (nullable string; populated on `Start` only)
-- `currentFrame`: Current replay frame index (nullable int)
-- `totalFrames`: Total frames in the recording (nullable int)
-- `progress`: Replay progress (nullable float in 0.0 – 1.0)
-- `isReplaying`: Whether replay is currently active (nullable bool)
+
+- `Success`: Whether the operation succeeded
+- `Message`: Status message
+- `Action`: Echoes which action was executed (`Start`, `Stop`, or `Status`)
+- `InputPath`: Path to recording file (nullable string; populated on `Start` only)
+- `CurrentFrame`: Current replay frame index (nullable int)
+- `TotalFrames`: Total frames in the recording (nullable int)
+- `Progress`: Replay progress (nullable float in 0.0 – 1.0)
+- `IsReplaying`: Whether replay is currently active (nullable bool)

.agents/skills/uloop-run-tests/SKILL.md

@@ -6,13 +6,13 @@ description: "Run Unity Test Runner and report detailed results. Use for EditMod
 
 # uloop run-tests
 
-Execute Unity Test Runner. When tests fail, NUnit XML results with error messages and stack traces are automatically saved. Read the XML file at `xmlPath` for detailed failure diagnosis.
+Execute Unity Test Runner. When tests fail, NUnit XML results with error messages and stack traces are automatically saved. Read the XML file at `XmlPath` for detailed failure diagnosis.
 
 Before running tests, run `uloop compile` first if you created, deleted, renamed, moved, or edited C# files, test files, `.asmdef`/`.asmref`, or package manifests since the last successful compile. This refreshes the AssetDatabase and surfaces compile errors before test execution.
 
-Before executing tests, `uloop run-tests` saves unsaved loaded Scene changes and unsaved current Prefab Stage changes by default. If saving fails, it returns `success: false`, keeps `testCount` at `0`, lists the unsaved items in `message`, and does not start the Unity Test Runner.
+Before executing tests, `uloop run-tests` saves unsaved loaded Scene changes and unsaved current Prefab Stage changes by default. If saving fails, it returns `Success: false`, keeps `TestCount` at `0`, lists the unsaved items in `Message`, and does not start the Unity Test Runner.
 
-`NoTestsFound` means zero tests matched — not a test failure. Check `noTestsFoundExplanation` and `message` for asmdef hints.
+`NoTestsFound` means zero tests matched — not a test failure. Check `NoTestsFoundExplanation` and `Message` for asmdef hints.
 
 ## Usage
 
@@ -57,22 +57,24 @@ uloop run-tests --filter-type regex --filter-value ".*Integration.*"
 ## Output
 
 Returns JSON with:
-- `success` (boolean): Whether all tests passed
-- `status` (string): Machine-readable execution status such as `Passed`, `Failed`, `NoTestsFound`, or `ExecutionFailed`
-- `hasFailures` (boolean): Whether any discovered test failed
-- `message` (string): Summary message
-- `noTestsFound` (boolean): Whether Unity Test Runner discovered zero matching tests
-- `noTestsFoundExplanation` (string): Agent-facing explanation when `noTestsFound` is true; empty otherwise
-- `completedAt` (string): ISO timestamp when the run finished
-- `testCount` (number): Total tests executed
-- `passedCount` (number): Passed tests
-- `failedCount` (number): Failed tests
-- `skippedCount` (number): Skipped tests
-- `xmlPath` (string): Path to NUnit XML result file. Empty string when no XML was saved (typically on `success: true`); populated only when tests failed and the XML file exists on disk.
+
+- `Success` (boolean): Whether all tests passed
+- `Status` (string): Machine-readable execution status such as `Passed`, `Failed`, `NoTestsFound`, or `ExecutionFailed`
+- `HasFailures` (boolean): Whether any discovered test failed
+- `Message` (string): Summary message
+- `NoTestsFound` (boolean): Whether Unity Test Runner discovered zero matching tests
+- `NoTestsFoundExplanation` (string): Agent-facing explanation when `NoTestsFound` is true; empty otherwise
+- `CompletedAt` (string): ISO timestamp when the run finished
+- `TestCount` (number): Total tests executed
+- `PassedCount` (number): Passed tests
+- `FailedCount` (number): Failed tests
+- `SkippedCount` (number): Skipped tests
+- `XmlPath` (string): Path to NUnit XML result file. Empty string when no XML was saved (typically on `Success: true`); populated only when tests failed and the XML file exists on disk.
 
 ### XML Result File
 
 When tests fail, NUnit XML results are automatically saved to `{project_root}/.uloop/outputs/TestResults/<timestamp>.xml`. The XML contains per-test-case results including:
+
 - Test name and full name
 - Pass/fail/skip status and duration
 - For failed tests: `<message>` (assertion error) and `<stack-trace>`