bluekiwi 0.1.7 → 0.2.0

package/dist/assets/skills/bk-rewind/SKILL.md

@@ -1,77 +1,70 @@
 ---
 name: bk-rewind
-description: BlueKiwi 특정 스텝으로 되돌아가기 스킬. 진행 중인 태스크의 이전 단계 또는 특정 단계로 되돌아갑니다. This skill should be used when the user says "/bk-rewind", "이전 단계", "다시 돌아가", "rewind", or wants to go back to a previous step in the BlueKiwi chain.
+description: BlueKiwi step rewind skill. Returns to a previous or specific step in a running task. This skill should be used when the user says "/bk-rewind", "go back", "previous step", "rewind", or wants to return to an earlier step in a BlueKiwi task.
 user_invocable: true
 ---
 
 # BlueKiwi Rewind
 
-진행 중인 태스크의 특정 스텝으로 되돌아가는 스킬. 이전 실행 기록은 보존되며, 해당 스텝에 새 pending 로그가 생성된다.
+Return to a specific step in a running task. Previous execution logs are preserved; a new pending log is created at the target step.
 
-## 인자 처리
+## Argument Handling
 
-- `/bk-rewind` → 인자 없음. `AskUserQuestion`으로 되돌아갈 스텝을 선택하도록 한다.
-- `/bk-rewind 3` → 스텝 번호 직접 지정.
-- `/bk-rewind 명확화 질문` → 스텝 제목으로 매칭.
+- `/bk-rewind` → No argument. Ask the user to select a step via AskUserQuestion.
+- `/bk-rewind 3` → Target step number provided directly.
+- `/bk-rewind clarification question` → Match by step title.
 
-## MCP 도구
+## Execution Steps
 
-- `mcp__bluekiwi__advance` (peek: true) — 현재 태스크/스텝 상태 확인
-- `mcp__bluekiwi__rewind` — 특정 스텝으로 되돌아가기
-- `mcp__bluekiwi__get_comments` — 코멘트 조회
+### 1. Verify Active Task
 
-## 실행 절차
+Call `advance` with `peek: true` to check the active task.
 
-### 1. 현재 태스크 확인
+- If no active task → show "No active task." and exit.
 
-`advance`를 `peek: true`로 호출하여 활성 태스크를 확인한다.
+### 2. Determine Target Step
 
-- 활성 태스크가 없으면 → "활성 태스크가 없습니다." 안내 후 종료.
+If no argument, show a selection UI via AskUserQuestion:
 
-### 2. 되돌아갈 스텝 결정
+- List completed steps as options
+- Include a summary of the step's output in each option's description
+- Show any comments on that step in the preview
 
-인자가 없으면 `AskUserQuestion`으로 선택 UI를 표시한다:
+If the argument is a number → use that step number directly.
+If the argument is text → match against step titles to find the step number.
 
-- 완료된 스텝 목록을 옵션으로 제공
-- 각 옵션의 description에 해당 스텝의 output 요약 포함
-- preview에 해당 스텝에 달린 코멘트 표시
+### 3. Confirm User Requirements
 
-인자가 숫자이면 해당 스텝 번호를 사용한다.
-인자가 텍스트이면 스텝 제목과 매칭하여 스텝 번호를 결정한다.
-
-### 3. 사용자 요구사항 확인
-
-되돌아가기 전에 `AskUserQuestion`으로 물어본다:
+Before rewinding, ask via AskUserQuestion:
 
 ```
-"Step [N]([제목])으로 되돌아갑니다. 이 단계를 다시 실행할 때 특별히 반영할 요구사항이 있으신가요?"
+"Going back to Step [N] ({title}). Do you have any specific requirements for this re-run?"
 ```
 
-옵션:
+Options:
 
-- "그대로 다시 진행 (Recommended)" — 추가 컨텍스트 없이 진행
-- "요구사항 있음" — Other로 자유 입력
+- "Proceed as-is (Recommended)" — continue without additional context
+- "I have specific requirements" — free text input via Other
 
-사용자가 요구사항을 입력하면 해당 내용을 코멘트로 서버에 저장한다 (task_comments).
+If the user enters requirements, save them as a comment on the task (task_comments).
 
-### 4. Rewind 실행
+### 4. Execute Rewind
 
-`mcp__bluekiwi__rewind`를 호출하여 해당 스텝으로 되돌아간다.
+Call `rewind` to go back to the target step.
 
-결과 표시:
+Show the result:
 
 ```
 ━━━━━━━━━━━━━━━━━━━━━━━━━
-Step [N]/[Total]로 되돌아갑니다: [노드 제목] [[node_type]]
-(이전 실행 기록은 보존됩니다)
+Going back to Step [N]/[Total]: [{node title}] [{node_type}]
+(Previous execution logs are preserved)
 ━━━━━━━━━━━━━━━━━━━━━━━━━
-[instruction 내용]
+Type `/bk-next` to re-execute this step.
 ━━━━━━━━━━━━━━━━━━━━━━━━━
-→ /bk-next 로 이 노드를 다시 실행하세요.
 ```
 
-## 중요 규칙
+## Rules
 
-- 되돌아가기 전에 항상 사용자에게 추가 요구사항이 있는지 확인할 것.
-- 요구사항이 있으면 코멘트로 서버에 저장하여 `/bk-next` 실행 시 참고되도록 할 것.
-- `AskUserQuestion`으로 모든 선택을 처리할 것.
+- Always ask the user for additional requirements before rewinding.
+- If requirements are provided, save them as a comment so `/bk-next` can reference them.
+- Handle all choices via AskUserQuestion.

package/dist/assets/skills/bk-run/SKILL.md

@@ -0,0 +1,50 @@
+---
+name: bk-run
+description: BlueKiwi workflow run skill. Selects a workflow and immediately starts it — a streamlined alias for /bk-start with explicit workflow selection. This skill should be used when the user says "/bk-run", "run workflow", "execute workflow", or wants to quickly start a specific workflow by name.
+user_invocable: true
+---
+
+# BlueKiwi Workflow Run
+
+Quickly select and start a specific workflow. Delegates to `/bk-start` execution flow once a workflow is selected.
+
+## Argument Handling
+
+- `/bk-run` → List workflows and ask the user to select.
+- `/bk-run <name>` → Find the best-matching workflow and confirm before starting.
+
+## Execution Steps
+
+### Step 1: Select Workflow
+
+Call `list_workflows` to get the active workflow list.
+
+If an argument is provided, find the best match by title and confirm:
+
+```
+Run '{title}' (v<version>)?
+```
+
+Ask via AskUserQuestion:
+
+- header: "Run workflow"
+- options: ["Run", "Pick a different one", "Cancel"]
+
+If no argument, show the full list via AskUserQuestion (up to 4 options).
+
+### Step 2: Start
+
+Call `start_workflow` with the selected `workflow_id`.
+
+Collect and pass `session_meta` exactly as described in `/bk-start`.
+
+### Step 3: Execute
+
+Continue with the same execution loop as `/bk-start`:
+
+- Show the roadmap.
+- Execute the first step immediately.
+- Follow the auto_advance loop.
+- Pause at `auto_advance: false` steps — if `execute_step` returns `next_action: "wait_for_human_approval"`, call `request_approval` and stop (HITL pause).
+
+Refer to `/bk-start` for all execution details including HITL handling, credential handling, and execute_step requirements.

package/dist/assets/skills/bk-scan/SKILL.md

@@ -0,0 +1,115 @@
+---
+name: bk-scan
+description: BlueKiwi compliance scan skill. Scans a local repository path for security and compliance risks using static pattern analysis, then saves findings to a BlueKiwi task. This skill should be used when the user says "/bk-scan", "scan for security issues", "compliance scan", or wants to run a repository scan with BlueKiwi.
+user_invocable: true
+---
+
+# BlueKiwi Compliance Scan
+
+Scan a local repository for security and compliance risks using built-in static patterns, then optionally link results to a BlueKiwi task.
+
+## Argument Handling
+
+- `/bk-scan` → Ask for the path to scan.
+- `/bk-scan <path>` → Start scanning the given path immediately.
+- `/bk-scan <path> --task <task_id>` → Scan and attach findings to a specific task.
+
+## Built-in Rule Set: korea-ota-code
+
+The default rule set checks for:
+
+| ID                | Severity | Description                                 |
+| ----------------- | -------- | ------------------------------------------- |
+| PIPA-001-RRN      | REVIEW   | Korean Resident Registration Number pattern |
+| ISMS-001-SECRET   | BLOCK    | Hardcoded secret/API key                    |
+| PIPA-002-FIELD    | REVIEW   | High-risk PII field names                   |
+| ISMS-004-HTTP     | WARN     | Plaintext HTTP URL (external)               |
+| LIA-001-GEO       | REVIEW   | Geolocation API usage                       |
+| PIPA-004-PRECHECK | REVIEW   | Pre-checked consent checkbox                |
+
+## Execution Steps
+
+### Step 1: Get Scan Path
+
+If no argument, ask via AskUserQuestion:
+
+- header: "Scan path"
+- "Enter the repository path to scan."
+- options: ["./ (current directory)", "src/", "Type a path"]
+
+### Step 2: Configure Options (Optional)
+
+Ask via AskUserQuestion:
+
+- header: "Scan options"
+- options: ["Default (korea-ota-code rules, 200 max matches)", "Custom settings"]
+
+If "Custom settings":
+
+- Max matches: ask for a number (default 200, max 1000)
+- Additional custom patterns: accept in `{ id, regex, description, severity }` format
+
+### Step 3: Run Scan
+
+Call `scan_repo`:
+
+```json
+{
+  "path": "<scan path>",
+  "rule_set": "korea-ota-code",
+  "max_matches": 200,
+  "task_id": <if provided>
+}
+```
+
+### Step 4: Display Results
+
+```
+Scan Results: {path}
+━━━━━━━━━━━━━━━━━━━━━━━━━
+Files scanned: {n}
+Matches found: {total}
+
+BLOCK (must fix before release): {count}
+  ISMS-001  src/config.ts:42   Hardcoded secret: API_KEY="sk-..."
+
+REVIEW (review recommended): {count}
+  PIPA-002  src/models/user.ts:15  High-risk field: residentRegistration
+
+WARN (best practice): {count}
+  ISMS-004  src/api/client.ts:8   Plaintext HTTP: http://external-api.com
+━━━━━━━━━━━━━━━━━━━━━━━━━
+```
+
+### Step 5: Save Findings (Optional)
+
+If a `task_id` was provided or an active task exists, ask via AskUserQuestion:
+
+- header: "Save findings?"
+- "Attach these findings to Task #{task_id}?"
+- options: ["Save findings", "Skip"]
+
+If saving, call `save_findings`:
+
+```json
+{
+  "task_id": <id>,
+  "findings": [
+    {
+      "rule_id": "ISMS-001",
+      "severity": "BLOCK",
+      "file": "src/config.ts",
+      "line": 42,
+      "description": "Hardcoded secret detected"
+    }
+  ]
+}
+```
+
+### Step 6: Remediation Guidance
+
+For each BLOCK finding, provide a brief remediation suggestion:
+
+- Hardcoded secrets → move to environment variables or a secrets manager
+- HTTP URLs → upgrade to HTTPS
+- RRN patterns → apply masking or tokenization

package/dist/assets/skills/bk-share/SKILL.md

@@ -0,0 +1,99 @@
+---
+name: bk-share
+description: BlueKiwi sharing skill. Shares a folder or workflow with a user group at a specified access level. This skill should be used when the user says "/bk-share", "share workflow", "share folder", or wants to share BlueKiwi content with a group.
+user_invocable: true
+---
+
+# BlueKiwi Share
+
+Share a folder with a user group at a specified access level (viewer or editor).
+
+## Argument Handling
+
+- `/bk-share` → Show resource and group selection.
+- `/bk-share <folder name>` → Pre-select the folder and ask for group/access level.
+
+## Core Principles
+
+- Sharing is applied at the **folder level**. Workflows and instructions inside the folder inherit the visibility.
+- Access levels: `viewer` (read-only) or `editor` (can modify contents).
+- Only the folder owner or an admin can share.
+
+## Execution Steps
+
+### Step 1: Select Resource
+
+Call `list_folders` to get the accessible folder list.
+
+Ask via AskUserQuestion:
+
+- header: "Share what?"
+- "Which folder would you like to share?"
+- options: folder name list (up to 4)
+
+### Step 2: Select Group
+
+Call `list_my_groups` to get groups the current user belongs to.
+
+Ask via AskUserQuestion:
+
+- header: "Share with"
+- "Which group should have access?"
+- options: group name list
+
+### Step 3: Set Access Level
+
+Ask via AskUserQuestion:
+
+- header: "Access level"
+- options: ["Viewer (read-only)", "Editor (can modify)"]
+
+### Step 4: Confirm and Apply
+
+Show a confirmation summary:
+
+```
+Share settings:
+━━━━━━━━━━━━━━━━━━━━━━━━━
+Folder: {folder name}
+Group:  {group name}
+Access: {viewer | editor}
+━━━━━━━━━━━━━━━━━━━━━━━━━
+```
+
+Ask via AskUserQuestion:
+
+- header: "Confirm"
+- options: ["Share", "Cancel"]
+
+Call `share_folder`:
+
+```json
+{
+  "folder_id": <id>,
+  "group_id": <id>,
+  "access_level": "viewer" | "editor"
+}
+```
+
+### Step 5: Result
+
+On success:
+
+```
+✅ Shared
+'{folder name}' is now accessible to '{group name}' as {access level}.
+
+To revoke access later, use `/bk-share` and select "Remove share".
+```
+
+## Revoke Access
+
+If the user says "remove share" or "unshare" → call `unshare_folder`:
+
+```json
+{
+  "folder_id": <id>,
+  "group_id": <id>
+}
+```

package/dist/assets/skills/bk-start/SKILL.md

@@ -1,63 +1,63 @@
 ---
 name: bk-start
-description: BlueKiwi 워크플로 실행 시작 스킬. 서버에 등록된 워크플로를 선택하고 첫 번째 단계를 즉시 실행합니다. This skill should be used when the user says "/bk-start", "워크플로 시작", "BlueKiwi 시작", or wants to begin a registered instruction workflow.
+description: BlueKiwi workflow execution skill. Selects a registered workflow and starts the first step immediately. This skill should be used when the user says "/bk-start", "start workflow", "run BlueKiwi", or wants to begin a registered instruction workflow.
 user_invocable: true
 ---
 
 # BlueKiwi Workflow Start
 
-서버에 등록된 워크플로를 선택하고, 태스크를 생성한 뒤, 첫 번째 단계를 즉시 실행하는 스킬.
+Select a registered workflow, create a task, and immediately execute the first step.
 
-## 인자 처리
+## Argument Handling
 
-- `/bk-start` → 워크플로 목록 조회 후 `AskUserQuestion`으로 선택 요청.
-- `/bk-start 보안 리뷰` → 주제 매칭된 워크플로를 Recommended로 제안.
+- `/bk-start` → Fetch workflow list, ask user to select via AskUserQuestion.
+- `/bk-start security review` → Propose the best-matching workflow as Recommended.
 
-## 핵심 원칙
+## Core Principles
 
-- **instruction은 에이전트의 내부 동작 지침이다. 사용자에게 그대로 노출하지 않는다.**
-- "노드", "node_type", "chain_nodes" 등 시스템 용어를 사용자에게 보여주지 않는다.
-- 사용자에게는 "단계" 또는 "스텝"으로만 안내한다.
-- output은 사용자가 웹 UI에서 읽는 기록이다. **"~했습니다"체로 통일**한다. "~함", "~함." 어투를 쓰지 않는다.
+- **Instructions are internal agent directives. Never expose raw instruction text to the user.**
+- Never use system terms like "node", "node_type", "chain_nodes" with the user.
+- Refer to steps as "step" only.
+- All `output` written to the server must be in past-tense declarative form ("I analyzed...", "I generated...").
 
-## AskUserQuestion 파라미터 규칙
+## AskUserQuestion Parameter Rules
 
-- `options`는 반드시 2~4개.
-- `preview`는 단순 문자열. 줄바꿈은 `\n`.
-- `header`는 12자 이내.
-- `multiSelect`는 `false`.
+- `options` must have 2–4 entries.
+- `preview` is a plain string. Use `\n` for line breaks.
+- `header` must be 12 characters or fewer.
+- `multiSelect` must be `false`.
 
-## 세션 복원 (미완료 태스크 이어가기)
+## Session Restore (Resume In-Progress Task)
 
-시작 전에 `curl -s http://localhost:3000/api/tasks`로 running 상태의 태스크가 있는지 확인한다.
+Before starting, check for running tasks using `advance(task_id, peek=true)` if a task ID is known, or prompt the user.
 
-미완료 태스크가 있으면 AskUserQuestion으로 물어본다:
+If an in-progress task is found, ask via AskUserQuestion:
 
-- header: "이어가기"
-- "진행 중인 태스크 #{id} ({워크플로명}, Step {N}/{total})가 있습니다. 이어서 진행하시겠습니까?"
-- 옵션: "이어서 진행 (Recommended)" / "새 워크플로 시작"
+- header: "Resume?"
+- "Task #{id} ({workflow name}, Step {N}/{total}) is in progress. Resume or start a new workflow?"
+- options: "Resume (Recommended)" / "Start new workflow"
 
-이어서 진행을 선택하면 → `advance(task_id, peek=true)` 호출 후 `task_context`를 읽고, `/bk-next` 스킬 로직으로 전환한다.
+If resuming → call `advance(task_id, peek=true)`, read `task_context`, then switch to `/bk-next` flow.
 
-## execute_step 호출 시 필수 파라미터
+## execute_step Required Parameters
 
 <HARD-RULE>
-execute_step 호출 시 반드시 아래 파라미터를 채운다:
-- `context_snapshot`: JSON 문자열. 이 단계에서 내린 결정사항, 주요 발견, 다음 단계 힌트.
-- `agent_id`: 사용 중인 모델명 (예: "claude-opus-4")
-- `user_name`: 사용자 이름 (알 수 없으면 생략 가능)
+Always populate these parameters when calling execute_step:
+- `context_snapshot`: JSON string. Store decisions made, key findings, and hints for the next step.
+- `agent_id`: Model name in use (e.g., "claude-opus-4-6")
+- `user_name`: User name (omit if unknown)
 
-파일을 생성하거나 수정한 경우, `artifacts` 배열에 기록한다:
+If files were created or modified, record them in the `artifacts` array:
 
-- 파일 생성: `{artifact_type: "file", title: "설계 문서", file_path: "docs/specs/design.md"}`
-- 커밋한 경우: `{artifact_type: "git_commit", title: "구현", git_ref: "커밋해시"}`
+- File created: `{artifact_type: "file", title: "Design Doc", file_path: "docs/specs/design.md"}`
+- Git commit: `{artifact_type: "git_commit", title: "Implementation", git_ref: "<hash>"}`
   </HARD-RULE>
 
-## 세션 메타 수집
+## Session Metadata Collection
 
 <HARD-RULE>
-`start_workflow` 호출 전에 반드시 세션 메타 정보를 수집하여 `session_meta` 파라미터로 전달한다.
-Bash 도구로 아래 명령을 실행하여 정보를 수집한다:
+Before calling `start_workflow`, collect session metadata and pass it as `session_meta`.
+Run the following via Bash:
 
 ```bash
 echo "PROJECT_DIR: $(pwd)"
@@ -67,7 +67,7 @@ echo "USER: $(whoami 2>/dev/null || echo 'unknown')"
 echo "OS: $(uname -s 2>/dev/null || echo 'unknown') $(uname -m 2>/dev/null)"
 ```
 
-수집한 정보를 JSON으로 구성한다:
+Build a JSON object:
 
 ```json
 {
@@ -82,61 +82,64 @@ echo "OS: $(uname -s 2>/dev/null || echo 'unknown') $(uname -m 2>/dev/null)"
 }
 ```
 
-- `agent`: 항상 "claude-code" (Claude Code 환경에서 실행 시)
-- `model_id`: 현재 사용 중인 모델 ID (system prompt에서 확인 가능)
-- `started_at`: 현재 UTC 시각
+- `agent`: always "claude-code" when running in Claude Code
+- `model_id`: current model ID (check system prompt)
+- `started_at`: current UTC time
   </HARD-RULE>
 
-## Credential 사용 (API 서비스 노드)
+## Credential Handling (API Service Nodes)
 
-advance 반환에 `credentials` 필드가 있으면 해당 노드는 외부 API 연동이 필요한 노드이다.
+If the `advance` response includes a `credentials` field, the node requires external API integration.
 
 <HARD-RULE>
-credentials.secrets의 키-값을 사용하여 API 호출을 수행한다.
-예: credentials.secrets.ACCESS_TOKEN → curl -H "Authorization: Bearer $TOKEN" 형태로 사용
-execute_step의 output에 시크릿 원본(토큰, 키 값)을 절대 포함하지 않는다.
-결과(URL, 상태코드, 응답 요약)만 기록한다.
+Use key-value pairs from `credentials.secrets` to make API calls.
+Example: `credentials.secrets.ACCESS_TOKEN` → `curl -H "Authorization: Bearer $TOKEN"`
+Never include raw secret values (tokens, keys) in `execute_step` output.
+Record only results (URL, status code, response summary).
 </HARD-RULE>
-## 실행 절차
 
-### 1. 워크플로 목록 조회 + 선택
+## Execution Steps
 
-`list_workflows`로 워크플로 목록을 가져온다.
+### 1. Fetch and Select Workflow
 
-**워크플로가 1개일 때** (#10): AskUserQuestion 대신 간단히 확인만 요청한다:
+Call `list_workflows` to retrieve the list.
 
-- "기능 브레인스토밍 워크플로를 시작합니다. 괜찮으신가요?" (AskUserQuestion으로 "시작/취소" 2개 옵션)
+**Single workflow**: Skip the selection UI, just confirm:
 
-**워크플로가 2개 이상일 때**: AskUserQuestion으로 선택 UI를 표시한다.
+- "Start the '{title}' workflow?" (AskUserQuestion: "Start" / "Cancel")
 
-### 2. 태스크 생성
+**Multiple workflows**: Show selection via AskUserQuestion.
 
-`start_workflow`을 호출한다. 인자가 있으면 `context`로 전달한다.
+### 2. Create Task
 
-### 3. 첫 번째 단계 즉시 실행 + auto_advance 루프
+Call `start_workflow`. Pass any argument as `context`.
 
-반환된 첫 번째 단계의 instruction을 **내부 지침으로 읽고 즉시 실행**한다.
+### 3. Execute First Step + auto_advance Loop
 
-실행 후 `execute_step`으로 결과를 저장하고, `advance`로 다음 단계를 확인한다.
+Read the first step's instruction as an **internal directive and execute immediately**.
 
-**로드맵 표시** (#20): 시작 시 전체 단계 흐름을 간략히 보여준다:
+After execution, save with `execute_step`, then check the response for `next_action` before calling `advance`.
+
+**Show roadmap at start** (#20):
 
 ```
-기능 브레인스토밍 시작 (11단계)
+Starting: {workflow title} ({n} steps)
 ━━━━━━━━━━━━━━━━━━━━━━━━━
-**1**→2→3→4→5→6→7→8→9→10→11
+**1** → 2 → 3 → 4 → 5 → 6 → 7
 ━━━━━━━━━━━━━━━━━━━━━━━━━
 ```
 
-**auto_advance 루프**: 다음 단계의 `auto_advance`가 `true`이면 멈추지 않고 계속 실행한다.
+**auto_advance loop**: If the next step has `auto_advance: true`, continue executing without pausing.
 
 <HARD-RULE>
-auto_advance=true인 단계를 실행한 후에는 반드시 다음 단계로 자동 진행한다.
-중간 결과를 간략히 표시한다: "✅ [제목] 완료 → 다음 단계 진행 중..."
-auto_advance=false인 단계를 만날 때까지 루프를 반복한다.
+After executing an auto_advance=true step, always proceed to the next step automatically.
+Show a brief inline update: "✅ [{title}] done → continuing to next step..."
+Repeat the loop until reaching an auto_advance=false step.
 </HARD-RULE>
 
-### 4. 멈출 때의 안내
+### 4. When Pausing
 
-- action 완료 후 (auto_advance=false): "다음 단계로 넘어가려면 `/bk-next`를 입력하세요."
-- gate 질문 표시 후: 사용자 응답을 기다린다. `/bk-next`를 안내하지 않는다.
+- **HITL** (execute_step returned `next_action: "wait_for_human_approval"`):
+  Call `request_approval`, show "⏸ Waiting for approval ��� use /bk-approve when ready.", stop.
+- After completing an action step (auto_advance=false, no HITL): "Type `/bk-next` to proceed."
+- After showing a gate question: Wait for user response. Do not show `/bk-next` hint.

package/dist/assets/skills/bk-status/SKILL.md

@@ -1,34 +1,34 @@
 ---
 name: bk-status
-description: BlueKiwi 태스크 상태 조회 스킬. 활성 태스크와 완료된 태스크의 진행 상황을 확인합니다. This skill should be used when the user says "/bk-status", "태스크 상태", "진행 상황", or wants to check BlueKiwi task status.
+description: BlueKiwi task status skill. Checks the progress of active and completed tasks. This skill should be used when the user says "/bk-status", "task status", "check progress", or wants to view BlueKiwi task status.
 user_invocable: true
 ---
 
 # BlueKiwi Task Status
 
-활성 태스크와 완료된 태스크의 진행 상황을 조회하는 스킬.
+View the progress of active and completed tasks.
 
-## 실행 절차
+## Execution Steps
 
-### 1. 태스크 조회
+### 1. Fetch Tasks
 
-`curl -s http://localhost:3000/api/tasks`로 태스크 목록을 조회한다.
+Call `advance` with `peek: true` on any known active task, or check the server for running tasks.
 
-### 2. 결과 표시
+### 2. Display Results
 
 ```
-BlueKiwi 태스크 현황:
+BlueKiwi Task Status
 ━━━━━━━━━━━━━━━━━━━━━━━━━
-  #1  [완료] 기능 브레인스토밍     8/8 steps   2026-04-07
-  #2  [실행중] PR 보안 리뷰       2/4 steps   2026-04-07  ← 활성
+  #1  [completed] Feature Brainstorm   8/8 steps   2026-04-07
+  #2  [running]   PR Security Review   2/4 steps   2026-04-07  ← active
 ━━━━━━━━━━━━━━━━━━━━━━━━━
 ```
 
-### 3. 상세 조회
+### 3. Detailed View
 
-인자로 태스크 번호가 주어지면 (`/bk-status 2`) 해당 태스크의 스텝별 로그를 표시한다.
+If a task number is given as an argument (`/bk-status 2`), show the step-by-step log for that task.
 
-## 참고
+## Notes
 
-- 웹 UI: http://localhost:3000/tasks
-- 실시간 모니터링: WebSocket Relay (`npm run ws`) 실행 필요
+- Web UI: http://localhost:3000/tasks
+- Real-time monitoring: Requires WebSocket Relay (`npm run ws`)