@markus-global/cli 0.9.0-rc.8 → 0.9.0

package/templates/roles/developer/HEARTBEAT.md

@@ -19,14 +19,14 @@
   - What coding patterns, debugging strategies, or decomposition approaches worked?
   - Did the reviewer leave feedback worth remembering?
   - Save insights via `memory_save` with `tags: ["insight", "coding"]` and `[INSIGHT]` format.
-  - If you found a repeatable workflow (e.g., "how to set up a new module", "how to debug integration tests"), promote it to MEMORY.md via `memory_update_longterm({ section: "procedures", ... })`.
+  - If you found a repeatable workflow (e.g., "how to set up a new module", "how to debug integration tests"), promote it to knowledge.md via `memory_update_longterm({ section: "procedures", ... })`.
   - When 3+ related insights accumulate, consider updating your ROLE.md with the new guideline.
 
 ## Self-Evolution
 
 - Reflect on what happened since last heartbeat. Save specific, actionable insights via `memory_save` with tags `["insight"]`. Format: `[INSIGHT] <summary>`. Examples: tool gotchas, better coding patterns, debugging shortcuts, error recovery strategies.
 - Check your revision rate — tasks with `executionRound > 1` needed revision. If revision rate is high (>30%), review whether saved insights cover the failure patterns.
-- Skip self-evolution if nothing meaningful happened.
+- Skip memory_save if nothing meaningful happened.
 
 ## Exit
 

package/templates/roles/devops/HEARTBEAT.md

@@ -19,7 +19,7 @@
   - What operational patterns or runbook steps worked well?
   - Were there deployment or rollback procedures worth documenting?
   - Save insights via `memory_save` with `tags: ["insight", "devops"]` and `[INSIGHT]` format.
-  - Promote repeatable runbooks to MEMORY.md via `memory_update_longterm({ section: "procedures", ... })`.
+  - Promote repeatable runbooks to knowledge.md via `memory_update_longterm({ section: "procedures", ... })`.
 
 ## Self-Evolution
 

package/templates/roles/finance/ROLE.md

@@ -69,7 +69,6 @@ When you receive a finance task, you follow a structured workflow:
 - **`spawn_subagent`**: Delegate parallel analysis of multiple department budgets, detailed transaction audits, or multi-entity financial consolidation.
 - **`web_search` / `web_fetch`**: Research tax regulation updates, accounting standard changes (GAAP/IFRS), and industry financial benchmarks.
 - **`deliverable_create` / `deliverable_search`**: Share financial reports, budget templates, and audit findings with stakeholders.
-- **`self-evolution`**: Continuously improve financial processes by capturing insights from each budget cycle and audit review.
 
 ## Quality Standards
 

package/templates/roles/finance/agent.json

@@ -6,9 +6,16 @@
   "description": "专业财务分析Agent — 预算编制、财务报表、费用审核、税务合规、财务风险管理",
   "author": "",
   "category": "management",
-  "tags": ["finance", "budgeting", "financial-reporting", "expense-audit", "tax-compliance", "risk-management"],
+  "tags": [
+    "finance",
+    "budgeting",
+    "financial-reporting",
+    "expense-audit",
+    "tax-compliance",
+    "risk-management"
+  ],
   "dependencies": {
-    "skills": ["self-evolution"],
+    "skills": [],
     "env": []
   },
   "agent": {

package/templates/roles/hr/ROLE.md

@@ -83,7 +83,6 @@ You use platform tools strategically across all HR workflows:
 - **`spawn_subagent`**: Delegate parallel candidate evaluations, survey analysis, or complex policy research to subagents for efficiency.
 - **`web_search` / `web_fetch`**: Research labor law updates, industry compensation benchmarks, best practice HR methodologies.
 - **`deliverable_create` / `deliverable_search`**: Share HR process templates, performance review formats, and onboarding documentation with the team.
-- **`self-evolution`**: Continuously improve your HR processes by capturing insights from each recruitment cycle and performance review season.
 
 ## Quality Standards
 

package/templates/roles/hr/agent.json

@@ -6,9 +6,17 @@
   "description": "专业人力资源Agent — 招聘流程管理、绩效评估、员工关系、入职离职管理、组织发展",
   "author": "",
   "category": "management",
-  "tags": ["hr", "recruitment", "performance-management", "employee-relations", "onboarding", "offboarding", "talent-management"],
+  "tags": [
+    "hr",
+    "recruitment",
+    "performance-management",
+    "employee-relations",
+    "onboarding",
+    "offboarding",
+    "talent-management"
+  ],
   "dependencies": {
-    "skills": ["self-evolution"],
+    "skills": [],
     "env": []
   },
   "agent": {

package/templates/roles/marketing/ROLE.md

@@ -70,7 +70,6 @@ When you receive a marketing task, you follow a structured workflow:
 - **`web_search` / `web_fetch`**: Research market trends, competitor strategies, channel benchmarks, and audience behavior data.
 - **`deliverable_create` / `deliverable_search`**: Share campaign plans, performance reports, content calendars, and competitive analysis.
 - **`humanizer`**: Apply to marketing copy to ensure natural, authentic-sounding communication that resonates with audiences.
-- **`self-evolution`**: Continuously improve marketing strategies by capturing insights from each campaign's performance data.
 
 ## Quality Standards
 

package/templates/roles/marketing/agent.json

@@ -6,9 +6,18 @@
   "description": "专业营销策略Agent — 营销策略制定、渠道分析、Campaign管理、用户增长、数据分析",
   "author": "",
   "category": "marketing",
-  "tags": ["marketing", "strategy", "campaign-management", "channel-analysis", "user-growth", "analytics"],
+  "tags": [
+    "marketing",
+    "strategy",
+    "campaign-management",
+    "channel-analysis",
+    "user-growth",
+    "analytics"
+  ],
   "dependencies": {
-    "skills": ["self-evolution", "humanizer"],
+    "skills": [
+      "humanizer"
+    ],
     "env": []
   },
   "agent": {

package/templates/roles/operations/ROLE.md

@@ -73,7 +73,6 @@ When you receive an operations task, you follow a structured workflow:
 - **`spawn_subagent`**: Delegate parallel process analysis for multiple departments, research industry best practices, analyze operational data sets.
 - **`web_search` / `web_fetch`**: Research operational best practices, industry benchmarks, regulatory requirements, and risk management frameworks.
 - **`deliverable_create` / `deliverable_search`**: Share process documentation, KPI dashboards, risk registers, and improvement proposals.
-- **`self-evolution`**: Capture lessons learned from each improvement cycle to refine your operations methodology.
 
 ## Quality Standards
 

package/templates/roles/operations/agent.json

@@ -6,9 +6,16 @@
   "description": "专业运营管理Agent — 流程优化、资源调配、KPI追踪、风险管理、运营数据分析",
   "author": "",
   "category": "management",
-  "tags": ["operations", "process-optimization", "resource-allocation", "kpi-tracking", "risk-management", "analytics"],
+  "tags": [
+    "operations",
+    "process-optimization",
+    "resource-allocation",
+    "kpi-tracking",
+    "risk-management",
+    "analytics"
+  ],
   "dependencies": {
-    "skills": ["self-evolution"],
+    "skills": [],
     "env": []
   },
   "agent": {

package/templates/roles/org-manager/HEARTBEAT.md

@@ -21,7 +21,7 @@
   - What management patterns worked? (task decomposition, reviewer assignment, workload balancing)
   - Did any workflow produce consistently good results?
   - Save insights via `memory_save` with `tags: ["insight", "management"]` and `[INSIGHT]` format.
-  - If you found a repeatable management workflow (e.g., "how to onboard a new agent", "how to handle cross-team blockers"), promote it to MEMORY.md via `memory_update_longterm({ section: "procedures", ... })`.
+  - If you found a repeatable management workflow (e.g., "how to onboard a new agent", "how to handle cross-team blockers"), promote it to knowledge.md via `memory_update_longterm({ section: "procedures", ... })`.
   - When 3+ related insights accumulate, consider updating your ROLE.md with the new guideline.
 - **Self-evolution**: Reflect on what happened since last heartbeat. Save specific, actionable insights via `memory_save` with tags `["insight"]`. Format: `[INSIGHT] <summary>`. Skip if nothing meaningful happened.
 - If nothing changed since last summary, respond HEARTBEAT_OK.

package/templates/roles/product-manager/HEARTBEAT.md

@@ -10,7 +10,7 @@
   - What product management patterns worked? (requirement writing, stakeholder alignment, sprint planning)
   - Were there requirements that completed without rework — what made them effective?
   - Save insights via `memory_save` with `tags: ["insight", "product"]` and `[INSIGHT]` format.
-  - If you found a repeatable PM workflow (e.g., "how to write effective requirements", "how to prioritize cross-team dependencies"), promote it to MEMORY.md via `memory_update_longterm({ section: "procedures", ... })`.
+  - If you found a repeatable PM workflow (e.g., "how to write effective requirements", "how to prioritize cross-team dependencies"), promote it to knowledge.md via `memory_update_longterm({ section: "procedures", ... })`.
   - When 3+ related insights accumulate, consider updating your ROLE.md with the new guideline.
 - **Self-evolution**: Reflect on what happened since last heartbeat. Save specific, actionable insights via `memory_save` with tags `["insight"]`. Format: `[INSIGHT] <summary>`. Examples: requirement patterns, stakeholder communication tips, prioritization insights. Skip if nothing meaningful happened.
 - If nothing changed since last summary, respond HEARTBEAT_OK.

package/templates/roles/project-manager/HEARTBEAT.md

@@ -19,7 +19,7 @@
   - What coordination or task-creation patterns led to smooth execution?
   - Were there blocker resolution or escalation approaches worth reusing?
   - Save insights via `memory_save` with `tags: ["insight", "project-management"]` and `[INSIGHT]` format.
-  - Promote repeatable PM workflows to MEMORY.md via `memory_update_longterm({ section: "procedures", ... })`.
+  - Promote repeatable PM workflows to knowledge.md via `memory_update_longterm({ section: "procedures", ... })`.
 
 ## Self-Evolution
 

package/templates/roles/qa-engineer/HEARTBEAT.md

@@ -18,7 +18,7 @@
   - What test strategies or reproduction techniques were effective?
   - Were there common defect patterns worth standardizing checks for?
   - Save insights via `memory_save` with `tags: ["insight", "testing"]` and `[INSIGHT]` format.
-  - Promote repeatable test workflows to MEMORY.md via `memory_update_longterm({ section: "procedures", ... })`.
+  - Promote repeatable test workflows to knowledge.md via `memory_update_longterm({ section: "procedures", ... })`.
 
 ## Self-Evolution
 

package/templates/roles/research-assistant/HEARTBEAT.md

@@ -18,7 +18,7 @@
   - What research methods or source evaluation approaches worked well?
   - Were there citation or synthesis patterns worth reusing?
   - Save insights via `memory_save` with `tags: ["insight", "research"]` and `[INSIGHT]` format.
-  - Promote repeatable research workflows to MEMORY.md via `memory_update_longterm({ section: "procedures", ... })`.
+  - Promote repeatable research workflows to knowledge.md via `memory_update_longterm({ section: "procedures", ... })`.
 
 ## Self-Evolution
 

package/templates/roles/reviewer/HEARTBEAT.md

@@ -14,7 +14,7 @@
   - What review patterns were effective? (checklist approaches, specific feedback formats, scope verification methods)
   - Were there common quality issues worth standardizing checks for?
   - Save insights via `memory_save` with `tags: ["insight", "review"]` and `[INSIGHT]` format.
-  - If you found a repeatable review workflow (e.g., "code review checklist for API changes", "review protocol for cross-team PRs"), promote it to MEMORY.md via `memory_update_longterm({ section: "procedures", ... })`.
+  - If you found a repeatable review workflow (e.g., "code review checklist for API changes", "review protocol for cross-team PRs"), promote it to knowledge.md via `memory_update_longterm({ section: "procedures", ... })`.
   - When 3+ related insights accumulate, consider updating your ROLE.md with the new guideline.
 - **Self-evolution**: Reflect on what happened since last heartbeat. Save specific, actionable insights via `memory_save` with tags `["insight"]`. Format: `[INSIGHT] <summary>`. Examples: common code issues spotted, review efficiency tips, quality patterns. Skip if nothing meaningful happened.
 - If nothing changed since last summary, respond HEARTBEAT_OK.

package/templates/roles/secretary/HEARTBEAT.md

@@ -56,7 +56,7 @@ Check `task_list` for tasks recently completed by you or your team members. For
 
 For identified best practices:
 - Save via `memory_save` with `tags: ["insight", "secretary"]` and `[INSIGHT]` format.
-- If it is a multi-step workflow (e.g., "new agent onboarding procedure", "org knowledge audit process"), promote it to MEMORY.md via `memory_update_longterm({ section: "procedures", ... })`.
+- If it is a multi-step workflow (e.g., "new agent onboarding procedure", "org knowledge audit process"), promote it to knowledge.md via `memory_update_longterm({ section: "procedures", ... })`.
 - When 3+ related insights accumulate, update `role/ROLE.md` under your agent home (absolute path in system context; read first via `file_read`, append only, log with `tags: ["insight", "role-evolution"]`). Do not write ROLE.md in the working directory.
 
 ## Self-Evolution Reflection

package/templates/roles/skill-architect/HEARTBEAT.md

@@ -18,7 +18,7 @@
   - What skill design patterns or instruction structures worked well?
   - Were there compatibility or testing approaches worth reusing?
   - Save insights via `memory_save` with `tags: ["insight", "skills"]` and `[INSIGHT]` format.
-  - Promote repeatable skill design workflows to MEMORY.md via `memory_update_longterm({ section: "procedures", ... })`.
+  - Promote repeatable skill design workflows to knowledge.md via `memory_update_longterm({ section: "procedures", ... })`.
 
 ## Self-Evolution
 

package/templates/roles/sre/ROLE.md

@@ -157,10 +157,6 @@ You leverage platform tools strategically across the incident lifecycle:
 - Alert dependent service owners about upstream/downstream outages
 - Share incident status with SRE team and management
 
-### self-evolution
-- Capture incident response patterns: what worked, what didn't, what to automate
-- Evolve runbook quality based on incidents handled
-- Improve triage accuracy over time through pattern learning
 
 ## Incident Command Structure
 

package/templates/roles/support/ROLE.md

@@ -74,7 +74,6 @@ When you receive a support task, you follow a structured workflow:
 - **`spawn_subagent`**: Delegate parallel research for complex technical issues, analyze customer satisfaction survey data.
 - **`web_search` / `web_fetch`**: Research product documentation, technical solutions, and industry best practices for support operations.
 - **`deliverable_create` / `deliverable_search`**: Share knowledge base articles, support metrics reports, and customer feedback analyses.
-- **`self-evolution`**: Continuously improve support processes by capturing insights from each ticket resolution and satisfaction survey.
 
 ## Quality Standards
 

package/templates/roles/support/agent.json

@@ -6,9 +6,16 @@
   "description": "专业客户支持Agent — 工单管理、SLA追踪、知识库维护、客户满意度管理、技术支持",
   "author": "",
   "category": "general",
-  "tags": ["support", "ticketing", "sla", "knowledge-base", "customer-satisfaction", "helpdesk"],
+  "tags": [
+    "support",
+    "ticketing",
+    "sla",
+    "knowledge-base",
+    "customer-satisfaction",
+    "helpdesk"
+  ],
   "dependencies": {
-    "skills": ["self-evolution"],
+    "skills": [],
     "env": []
   },
   "agent": {

package/templates/roles/tech-writer/HEARTBEAT.md

@@ -18,7 +18,7 @@
   - What documentation structures or explanation patterns worked well?
   - Were there code-to-docs verification workflows worth reusing?
   - Save insights via `memory_save` with `tags: ["insight", "documentation"]` and `[INSIGHT]` format.
-  - Promote repeatable documentation workflows to MEMORY.md via `memory_update_longterm({ section: "procedures", ... })`.
+  - Promote repeatable documentation workflows to knowledge.md via `memory_update_longterm({ section: "procedures", ... })`.
 
 ## Self-Evolution
 

package/templates/skills/cursor-agent/SKILL.md

@@ -141,7 +141,7 @@ invoke_coding_tool({
 ```
 invoke_coding_tool({
   tool: "cursor-agent",
-  prompt: "Add the coding-tools skill templates under templates/skills/. Follow the existing skill.json + SKILL.md pattern from templates/skills/self-evolution/.",
+  prompt: "Add the coding-tools skill templates under templates/skills/. Follow the existing skill.json + SKILL.md pattern from templates/skills/Learning Habits/.",
   workdir: "/path/to/markus-monorepo",
   task_id: "task-789"
 })

package/templates/skills/skill-building/SKILL.md

@@ -161,7 +161,7 @@ If the user only gives a Chinese title, **you invent an English kebab slug**, se
 - **`skillFile`**: Always `"SKILL.md"` — the entry point instruction document
 - **`requiredPermissions`**: (optional) Array of permissions: `"shell"`, `"file"`, `"network"`, `"browser"`
 - **`mcpServers`**: (optional) Map of MCP server name → config. Each config has `command`, `args?`, `env?`. Use `${SKILL_DIR}` in args/env to reference the skill directory.
-- **`alwaysOn`**: (optional, boolean) If `true`, the skill's instructions are automatically injected into **every** agent's context at startup. Only use this for foundational skills that all agents must always follow (e.g., `self-evolution`). Default is `false` — non-alwaysOn built-in skills are listed as "available" in the agent's identity section and can be activated on-demand via `discover_tools`.
+- **`alwaysOn`**: (optional, boolean) If `true`, the skill is listed in every agent's discoverable catalog as a foundational skill (e.g., Learning Habits). Full instructions are **not** auto-injected — agents activate via `discover_tools({ name: [...] })` (Hermes progressive disclosure). Default is `false`.
 
 ## After Creation
 

package/templates/teams/content-team/team.json

@@ -6,7 +6,18 @@
   "description": "A multi-platform content creation team with specialized writers for Chinese and English markets. Editor-in-Chief plans the editorial calendar and reviews content; Chinese writer handles WeChat public accounts and Zhihu long-form articles; Little Red Book (Xiaohongshu) operator creates image-text notes and short videos; English writer produces X.com threads and LinkedIn articles; General content creator handles multi-platform adaptation and asset reuse. Each writer owns their platform domain. Designed for content operations, brand marketing, and cross-platform content strategies.",
   "author": "Markus Team",
   "category": "productivity",
-  "tags": ["content", "writing", "marketing", "multi-platform", "social-media", "wechat", "xiaohongshu", "x.com", "linkedin", "zhihu"],
+  "tags": [
+    "content",
+    "writing",
+    "marketing",
+    "multi-platform",
+    "social-media",
+    "wechat",
+    "xiaohongshu",
+    "x.com",
+    "linkedin",
+    "zhihu"
+  ],
   "icon": "edit",
   "starterTasks": [
     {
@@ -46,7 +57,7 @@
         "roleName": "project-manager",
         "count": 1,
         "description": "选题策划、内容审核、发布排期协调",
-        "skills": ["self-evolution"]
+        "skills": []
       },
       {
         "name": "Chinese Writer",
@@ -54,7 +65,7 @@
         "roleName": "content-writer",
         "count": 1,
         "description": "微信公众号/知乎长文写作",
-        "skills": ["self-evolution"]
+        "skills": []
       },
       {
         "name": "Xiaohongshu Operator",
@@ -62,7 +73,10 @@
         "roleName": "content-writer",
         "count": 1,
         "description": "小红书图文/短视频创作与发布",
-        "skills": ["chrome-devtools", "xhs-posting", "self-evolution"]
+        "skills": [
+          "chrome-devtools",
+          "xhs-posting"
+        ]
       },
       {
         "name": "English Writer",
@@ -70,7 +84,9 @@
         "roleName": "content-writer",
         "count": 1,
         "description": "X.com/LinkedIn 英文推文与文章",
-        "skills": ["x-com-browser-posting", "self-evolution"]
+        "skills": [
+          "x-com-browser-posting"
+        ]
       },
       {
         "name": "General Content Creator",
@@ -78,9 +94,11 @@
         "roleName": "content-writer",
         "count": 1,
         "description": "多平台适应、素材复用、抖音口播文案",
-        "skills": ["self-evolution"]
+        "skills": []
       }
     ],
-    "workflows": ["workflows/content-publishing.yaml"]
+    "workflows": [
+      "workflows/content-publishing.yaml"
+    ]
   }
 }

package/templates/teams/dev-squad/NORMS.md

@@ -59,7 +59,7 @@ This is the most important rule for parallel development:
 
 - Document non-obvious decisions as `deliverable_create` (type: "architecture_decision").
 - Save reusable patterns and gotchas via `memory_save` with appropriate tags.
-- After completing a complex task, record lessons learned during the self-evolution reflection.
+- After completing a complex task, record lessons learned during the Learning Habits reflection.
 
 ## Quality Gates by Phase
 

package/templates/teams/dev-squad/team.json

@@ -6,7 +6,15 @@
   "description": "A high-performance development team that follows a structured Plan → Implement → Review → Validate workflow. Features parallel implementation with worktree isolation, two-stage review (code + QA), subagent-driven deep analysis, and dependency-aware task scheduling. Suitable for feature sprints, refactoring campaigns, and production-grade software delivery.",
   "author": "Markus Team",
   "category": "development",
-  "tags": ["development", "agile", "sprint", "feature", "software", "tdd", "code-review"],
+  "tags": [
+    "development",
+    "agile",
+    "sprint",
+    "feature",
+    "software",
+    "tdd",
+    "code-review"
+  ],
   "icon": "users",
   "i18n": {
     "zh-CN": {
@@ -28,35 +36,46 @@
         "role": "manager",
         "roleName": "project-manager",
         "count": 1,
-        "skills": ["team-building"]
+        "skills": [
+          "team-building"
+        ]
       },
       {
         "name": "Backend Developer",
         "role": "worker",
         "roleName": "developer",
         "count": 1,
-        "skills": ["self-evolution", "coding-tools"]
+        "skills": [
+          "coding-tools"
+        ]
       },
       {
         "name": "Frontend Developer",
         "role": "worker",
         "roleName": "developer",
         "count": 1,
-        "skills": ["chrome-devtools", "self-evolution", "coding-tools"]
+        "skills": [
+          "chrome-devtools",
+          "coding-tools"
+        ]
       },
       {
         "name": "Code Reviewer",
         "role": "worker",
         "roleName": "reviewer",
         "count": 1,
-        "skills": ["self-evolution", "coding-tools"]
+        "skills": [
+          "coding-tools"
+        ]
       },
       {
         "name": "QA Engineer",
         "role": "worker",
         "roleName": "qa-engineer",
         "count": 1,
-        "skills": ["self-evolution", "coding-tools"]
+        "skills": [
+          "coding-tools"
+        ]
       }
     ]
   }

package/templates/teams/engineering-pod/team.json

@@ -6,7 +6,14 @@
   "description": "A full-featured engineering pod with an architect who designs, multiple developers who implement in parallel across layers (backend, frontend, infrastructure), a dedicated reviewer, and a DevOps engineer. Designed for complex multi-layer projects requiring cross-cutting coordination, API contracts, and production deployment. Each member owns a clear domain to prevent file conflicts.",
   "author": "Markus Team",
   "category": "development",
-  "tags": ["engineering", "full-stack", "architecture", "devops", "deployment", "multi-layer"],
+  "tags": [
+    "engineering",
+    "full-stack",
+    "architecture",
+    "devops",
+    "deployment",
+    "multi-layer"
+  ],
   "icon": "server",
   "i18n": {
     "zh-CN": {
@@ -28,35 +35,46 @@
         "role": "manager",
         "roleName": "project-manager",
         "count": 1,
-        "skills": ["team-building", "self-evolution"]
+        "skills": [
+          "team-building"
+        ]
       },
       {
         "name": "Backend Engineer",
         "role": "worker",
         "roleName": "developer",
         "count": 1,
-        "skills": ["self-evolution", "coding-tools"]
+        "skills": [
+          "coding-tools"
+        ]
       },
       {
         "name": "Frontend Engineer",
         "role": "worker",
         "roleName": "developer",
         "count": 1,
-        "skills": ["chrome-devtools", "self-evolution", "coding-tools"]
+        "skills": [
+          "chrome-devtools",
+          "coding-tools"
+        ]
       },
       {
         "name": "Infra Engineer",
         "role": "worker",
         "roleName": "devops",
         "count": 1,
-        "skills": ["self-evolution", "coding-tools"]
+        "skills": [
+          "coding-tools"
+        ]
       },
       {
         "name": "Senior Reviewer",
         "role": "worker",
         "roleName": "reviewer",
         "count": 1,
-        "skills": ["self-evolution", "coding-tools"]
+        "skills": [
+          "coding-tools"
+        ]
       }
     ]
   }

package/templates/teams/research-lab/team.json

@@ -6,7 +6,16 @@
   "description": "A structured think tank research team. The Director defines research directions and assigns tasks; Senior Researcher leads deep investigations and produces reports; Junior Researcher handles data collection, preliminary analysis and literature review; Coordinator manages schedules, meeting notes, and deliverable organization. Designed for deep research projects, competitive analysis, industry reports, and knowledge synthesis. Features parallel investigation tracks, adversarial hypothesis testing, and evidence-based synthesis.",
   "author": "Markus Team",
   "category": "research",
-  "tags": ["research", "think-tank", "investigation", "analysis", "report", "knowledge", "synthesis", "deep-research"],
+  "tags": [
+    "research",
+    "think-tank",
+    "investigation",
+    "analysis",
+    "report",
+    "knowledge",
+    "synthesis",
+    "deep-research"
+  ],
   "icon": "search",
   "starterTasks": [
     {
@@ -50,7 +59,7 @@
         "roleName": "project-manager",
         "count": 1,
         "description": "制定调研方向、分配任务、审核综合产出",
-        "skills": ["self-evolution"]
+        "skills": []
       },
       {
         "name": "Senior Researcher",
@@ -58,7 +67,7 @@
         "roleName": "research-assistant",
         "count": 1,
         "description": "深度调研、报告撰写、方法论证",
-        "skills": ["self-evolution"]
+        "skills": []
       },
       {
         "name": "Junior Researcher",
@@ -66,7 +75,7 @@
         "roleName": "research-assistant",
         "count": 1,
         "description": "资料收集、初步分析、文献综述",
-        "skills": ["self-evolution"]
+        "skills": []
       },
       {
         "name": "Coordinator",
@@ -74,7 +83,7 @@
         "roleName": "operations",
         "count": 1,
         "description": "会议记录、进度跟踪、交付物整理",
-        "skills": ["self-evolution"]
+        "skills": []
       }
     ]
   }

package/templates/teams/startup-team/team.json

@@ -6,7 +6,15 @@
   "description": "A lean startup team built for speed. Product Manager drives strategy, Full-Stack Dev ships features rapidly, Growth Lead handles marketing and analytics, and DevOps keeps everything running. Each member owns their domain end-to-end. Designed for MVP development, rapid iteration, and growth experiments with subagent-assisted research and background deployment.",
   "author": "Markus Team",
   "category": "startup",
-  "tags": ["startup", "lean", "mvp", "product", "growth", "iteration", "agile"],
+  "tags": [
+    "startup",
+    "lean",
+    "mvp",
+    "product",
+    "growth",
+    "iteration",
+    "agile"
+  ],
   "icon": "zap",
   "i18n": {
     "zh-CN": {
@@ -26,21 +34,24 @@
         "role": "manager",
         "roleName": "product-manager",
         "count": 1,
-        "skills": ["self-evolution"]
+        "skills": []
       },
       {
         "name": "Full-Stack Developer",
         "role": "worker",
         "roleName": "developer",
         "count": 2,
-        "skills": ["chrome-devtools", "self-evolution", "coding-tools"]
+        "skills": [
+          "chrome-devtools",
+          "coding-tools"
+        ]
       },
       {
         "name": "Growth Lead",
         "role": "worker",
         "roleName": "marketing",
         "count": 1,
-        "skills": ["self-evolution"]
+        "skills": []
       }
     ]
   }