[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6DGVKHPJPVTTFNV3YTN085C":3,"skill-same:01M6DGVKHPJPVTTFNV3YTN085C":235},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":16,"categories":17,"keywords":23,"publishedAt":31,"linkOnly":32,"markdownContent":33,"markdownBody":34,"frontMatter":35,"refName":36,"commitSha":37,"skillDirectory":38,"forksCount":14,"repositoryHtmlUrl":39,"repositoryHomepageUrl":6,"ownerHtmlUrl":40,"ownerBlogUrl":41,"ownerDescription":42,"updatedAt":43,"install":44,"resources":57},"01M6DGVKHPJPVTTFNV3YTN085C","nexau-agent",null,"Guide for building NexAU agents from scratch. This skill should be used when implementing a new NexAU agent — including YAML configuration, system prompt, tool definitions, tool bindings, entry point scripts.","CharmCheen\u002Fschedule_assistant_based_on_NexAU","CharmCheen","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F152878977?v=4","01M6DGTZ0AQKZGTVZA595NTQBJ","Changhao Qin","01M6DGTZ7HV1TWSQENZBKDWMD2",0,"Apache-2.0","https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002FSKILL.md",[18],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000023","Skill与智能体开发","SCENARIO","使用场景",[24,25,26,27,28,29,30],"nexau agent","yaml","system","tool","prompt","guide","configuration","2026-11-10 10:12:30",false,"---\nname: nexau-agent\ndescription: Guide for building NexAU agents from scratch. This skill should be used when implementing a new NexAU agent — including YAML configuration, system prompt, tool definitions, tool bindings, entry point scripts.\n---\n\n# NexAU Agent Builder\n\nThis skill provides the procedural knowledge and reference material needed to implement NexAU agents correctly and efficiently.\n\n## When to Use\n\n- Implementing a new standalone agent (YAML config + system prompt + tools + entry point)\n- Adding custom tools to an existing agent\n- Writing or refining system prompts for NexAU agents\n- Setting up skills for an agent\n- Looking up NexAU framework concepts (agents, tools, LLMs, transports, sessions, etc.)\n\n> **NexAU Documentation**: The `docs\u002F` directory contains the full user-facing documentation for NexAU. Start with `docs\u002Findex.md` for an overview, `docs\u002Fgetting-started.md` for setup, `docs\u002Fcore-concepts\u002F` for fundamentals (agents, tools, LLMs), and `docs\u002Fadvanced-guides\u002F` for topics like skills, hooks, MCP, sandbox, transports, and session management.\n\n## Agent Implementation Workflow\n\n### 1. Understand the Requirements\n\nBefore writing any code, read the RFC or requirements document thoroughly. Identify:\n\n- Agent purpose and capabilities\n- Required tools (builtin vs custom)\n- LLM configuration needs (model, temperature, token limits)\n- Any skills the agent should have\n\n### 2. Create the Directory Structure\n\nA standalone agent follows this layout:\n\n```\nagent_name\u002F\n├── agent_name.yaml          # Agent configuration\n├── systemprompt.md           # System prompt (Jinja2 or plain string)\n├── start.py                  # Entry point script\n├── tools\u002F                    # Tool definitions\n│   ├── read_file.tool.yaml\n│   ├── write_file.tool.yaml\n│   └── custom_tool.tool.yaml\n├── skills\u002F                   # Optional skills\n│   └── skill-name\u002F\n│       └── SKILL.md\n└── custom_tools\u002F             # Custom tool implementations (if any)\n    └── custom_tool.py\n```\n\n\n### 3. Write the Agent YAML Config\n\nRefer to `references\u002Fagent-yaml-reference.md` for the complete field reference. Key points:\n\n- Use `${env.VAR_NAME}` for environment variable substitution (LLM keys, sandbox paths)\n- Set `system_prompt_type: jinja` when using template variables like `{{ date }}`\n- Set `system_prompt_type: string` for plain text prompts\n- Choose `tool_call_mode: openai` for OpenAI-compatible APIs\n- Reference tool YAMLs with relative paths: `.\u002Ftools\u002Ftool.tool.yaml`\n- Reference system prompt with relative path: `.\u002Fsystemprompt.md`\n\nTemplate available at: `assets\u002Ftemplates\u002Fagent.yaml`\n\n### 4. Define Tools\n\nFor each tool the agent needs:\n\n1. Check if a builtin tool exists — see `references\u002Fbuiltin-tools-reference.md`\n2. If builtin, copy the tool YAML from an existing example and set the binding path\n3. If custom, create both a `.tool.yaml` definition and a Python implementation\n\nRefer to `references\u002Ftool-yaml-reference.md` for the tool YAML schema.\n\nTemplate available at: `assets\u002Ftemplates\u002Ftool.tool.yaml`\n\n### 5. Write the System Prompt\n\nFollow the conventions in `references\u002Fsystem-prompt-guide.md`:\n\n- Start with a clear role description\n- Define a numbered step-by-step workflow\n- Include guidelines and constraints\n- Add template variables at the bottom: `{{ date }}`, `{{ username }}`, `{{ working_directory }}`\n\nTemplate available at: `assets\u002Ftemplates\u002Fsystemprompt.md`\n\n### 6. Create the Entry Point\n\nFor standalone agents, use `Agent.from_yaml()`.\n\nTemplate available at: `assets\u002Ftemplates\u002Fstart.py` (standalone)\n\n### 7. Verify the Implementation\n\n- **Validate agent config**:\n  ```bash\n  python {THIS_SKILL_FOLDER}\u002Fscripts\u002Fvalidate_agent.py path\u002Fto\u002Fagent.yaml\n  ```\n  - Add `--recursive` to also validate sub-agent configs\n  - Add `--json` for machine-readable output\n- Check Python syntax: `python -c \"import ast; ast.parse(open('start.py').read())\"`\n- Check imports: `python -c \"from nexau import Agent\"`\n- Run the agent with a simple test query\n\n## Common Patterns\n\n### Environment Variable Substitution in YAML\n\n```yaml\nllm_config:\n  model: ${env.LLM_MODEL}\n  base_url: ${env.LLM_BASE_URL}\n  api_key: ${env.LLM_API_KEY}\n```\n\n### Adding Builtin Tools\n\n```yaml\ntools:\n  - name: read_file\n    yaml_path: .\u002Ftools\u002Fread_file.tool.yaml\n    binding: nexau.archs.tool.builtin.file_tools:read_file\n```\n\n### Adding Skills\n\n```yaml\nskills:\n  - .\u002Fskills\u002Fskill-name\n```\n\n### Stop Tools\n\nUse `stop_tools` to define tools that halt the agent loop when called:\n\n```yaml\nstop_tools: [complete_task]    # For task-based agents\nstop_tools: [ask_user]         # For interactive agents\n```\n\n### Tracers\n\n```yaml\ntracers:\n  - import: nexau.archs.tracer.adapters.in_memory:InMemoryTracer\n```\n\n## Builtin Tools Catalog\n\nAll builtin tool YAML configs are located under `builtin_tools\u002Ftools\u002F` in this skill folder. When adding a builtin tool to an agent, copy the YAML to your agent's `tools\u002F` directory and set the binding path accordingly.\n\n### File Tools\n\n| Tool | Binding | Config YAML |\n|------|---------|-------------|\n| read_file | `nexau.archs.tool.builtin.file_tools:read_file` | `builtin_tools\u002Ftools\u002Fread_file.tool.yaml` |\n| write_file | `nexau.archs.tool.builtin.file_tools:write_file` | `builtin_tools\u002Ftools\u002Fwrite_file.tool.yaml` |\n| replace | `nexau.archs.tool.builtin.file_tools:replace` | `builtin_tools\u002Ftools\u002Freplace.tool.yaml` |\n| search_file_content | `nexau.archs.tool.builtin.file_tools:search_file_content` | `builtin_tools\u002Ftools\u002Fsearch_file_content.tool.yaml` |\n| glob | `nexau.archs.tool.builtin.file_tools:glob` | `builtin_tools\u002Ftools\u002FGlob.tool.yaml` |\n| list_directory | `nexau.archs.tool.builtin.file_tools:list_directory` | `builtin_tools\u002Ftools\u002Flist_directory.tool.yaml` |\n\n\n### Web Tools\n\n| Tool | Binding | Config YAML |\n|------|---------|-------------|\n| google_web_search | `nexau.archs.tool.builtin.web_tools:google_web_search` | `builtin_tools\u002Ftools\u002FWebSearch.tool.yaml` |\n| web_fetch | `nexau.archs.tool.builtin.web_tools:web_fetch` | `builtin_tools\u002Ftools\u002FWebFetch.tool.yaml` |\n\n### Shell Tools\n\n| Tool | Binding | Config YAML |\n|------|---------|-------------|\n| run_shell_command (sync) | `nexau.archs.tool.builtin.shell_tools:run_shell_command` | `builtin_tools\u002Ftools\u002Frun_shell_command_sync.tool.yaml` |\n\n### Session Tools\n\n| Tool | Binding | Config YAML |\n|------|---------|-------------|\n| write_todos | `nexau.archs.tool.builtin.session_tools:write_todos` | `builtin_tools\u002Ftools\u002Fwrite_todos.tool.yaml` |\n| complete_task | `nexau.archs.tool.builtin.session_tools:complete_task` | `builtin_tools\u002Ftools\u002Fcomplete_task.tool.yaml` |\n| ask_user | `nexau.archs.tool.builtin.session_tools:ask_user` | `builtin_tools\u002Ftools\u002Fask_user.tool.yaml` |\n\n\n> For detailed parameter descriptions of each tool, see `references\u002Fbuiltin-tools-reference.md`.\n\n## Bundled Resources\n\n| Path | Purpose |\n|------|---------|\n| `references\u002Fagent-yaml-reference.md` | Complete agent YAML config field reference |\n| `references\u002Ftool-yaml-reference.md` | Tool YAML definition schema and examples |\n| `references\u002Fbuiltin-tools-reference.md` | Catalog of all builtin tools with binding paths |\n| `references\u002Fsystem-prompt-guide.md` | System prompt writing conventions and patterns |\n| `assets\u002Ftemplates\u002Fagent.yaml` | Starter agent YAML config template |\n| `assets\u002Ftemplates\u002Ftool.tool.yaml` | Starter tool YAML definition template |\n| `assets\u002Ftemplates\u002Fsystemprompt.md` | Starter system prompt template |\n| `assets\u002Ftemplates\u002Fstart.py` | Standalone agent entry point template |\n| `scripts\u002Fvalidate_agent.py` | Agent YAML validator (schema + file reference checks) |\n| `docs\u002Findex.md` | NexAU documentation index — entry point for all user-facing docs |\n| `docs\u002Fgetting-started.md` | Installation, setup, and first agent walkthrough |\n| `docs\u002Fcore-concepts\u002Fagents.md` | Agent architecture, lifecycle, and configuration |\n| `docs\u002Fcore-concepts\u002Ftools.md` | Tool system — definition, binding, execution |\n| `docs\u002Fcore-concepts\u002Fllms.md` | LLM provider configuration and aggregator usage |\n| `docs\u002Fadvanced-guides\u002Fskills.md` | Skill system — folder-based and tool-based skills |\n| `docs\u002Fadvanced-guides\u002Fhooks.md` | Hook system for lifecycle event handling |\n| `docs\u002Fadvanced-guides\u002Fmcp.md` | Model Context Protocol integration |\n| `docs\u002Fadvanced-guides\u002Fsandbox.md` | Sandbox configuration and isolation |\n| `docs\u002Fadvanced-guides\u002Fsession-management.md` | Session persistence and history management |\n| `docs\u002Fadvanced-guides\u002Fstreaming-events.md` | Streaming event types and handling |\n| `docs\u002Fadvanced-guides\u002Ftransports.md` | Transport layer — HTTP, stdio, WebSocket, gRPC |\n| `docs\u002Fadvanced-guides\u002Ftemplating.md` | Jinja2 templating in system prompts and configs |\n| `docs\u002Fadvanced-guides\u002Fcontext_compaction.md` | Context compaction middleware for long conversations |\n| `docs\u002Fadvanced-guides\u002Fglobal-storage.md` | Global storage for cross-turn agent state |\n| `docs\u002Fadvanced-guides\u002Ftracer.md` | Tracing and observability setup |\n| `docs\u002Fadvanced-guides\u002Fimage.md` | Image input handling |\n\n## Setting Up Skills for an Agent\n\nSkills are reusable capabilities that extend an agent's knowledge and workflows. NexAU supports two types: **folder-based skills** and **tool-based skills**. Both are automatically registered and discoverable via the `LoadSkill` tool at runtime.\n\n### Folder-Based Skills\n\nA folder-based skill is a self-contained directory with a `SKILL.md` file. To add one to an agent:\n\n1. Create the skill directory under the agent's `skills\u002F` folder:\n\n```\nskills\u002F\n└── my-skill\u002F\n    ├── SKILL.md              # Required — metadata + instructions\n    ├── scripts\u002F              # Optional — executable code\n    ├── references\u002F           # Optional — context docs loaded on demand\n    └── assets\u002F               # Optional — templates, images, output files\n```\n\n2. Reference the skill in the agent YAML:\n\n```yaml\nskills:\n  - .\u002Fskills\u002Fmy-skill\n```\n\nThe agent will see a brief description of the skill in its system prompt and can call `LoadSkill` to read the full `SKILL.md` content when needed.\n\n### Tool-Based Skills\n\nA tool-based skill is a regular tool marked with `as_skill: true` in its YAML definition. This is useful for simple, well-defined capabilities that do not need extensive documentation.\n\n```yaml\n# tools\u002Fgenerate_code.tool.yaml\ntype: tool\nname: generate_code\ndescription: Generates code based on specifications\nas_skill: true\nskill_description: Code generation skill for multiple programming languages\ninput_schema:\n  type: object\n  properties:\n    language:\n      type: string\n    specification:\n      type: string\n  required: [language, specification]\n```\n\nThe `skill_description` field is required when `as_skill: true`. It provides the brief text shown in the skill registry.\n\n### Combining Both Types\n\nBoth skill types can coexist in the same agent:\n\n```yaml\n# Agent YAML\nskills:\n  - .\u002Fskills\u002Fdata-analysis       # Folder-based skill\n  - .\u002Fskills\u002Freport-writing      # Folder-based skill\n\ntools:\n  - name: web_search\n    yaml_path: .\u002Ftools\u002Fweb_search.tool.yaml   # Tool-based skill (as_skill: true)\n    binding: nexau.archs.tool.builtin.web_tools:google_web_search\n```\n\n### How Skills Work at Runtime\n\n1. All skills are registered in the agent's `skill_registry` during initialization\n2. A `LoadSkill` tool is automatically added to the agent\n3. The agent sees brief descriptions of all available skills in its system prompt\n4. When the agent needs detailed information, it calls `LoadSkill` with the skill name\n5. The full `SKILL.md` content (and skill folder path) is returned to the agent\n\n## Writing High-Quality Skills\n\nA well-crafted skill transforms a general-purpose agent into a specialized one. Follow these principles to maximize effectiveness.\n\n### SKILL.md Structure\n\nEvery `SKILL.md` must start with YAML frontmatter containing `name` and `description`:\n\n```markdown\n---\nname: my-skill\ndescription: Brief description of what this skill does. This skill should be used when [trigger conditions].\n---\n\n# Skill Title\n\n[Purpose — what this skill provides, in 1-2 sentences]\n\n## When to Use\n\n[Concrete trigger conditions — when should the agent load this skill]\n\n## Workflow \u002F Instructions\n\n[Step-by-step procedural knowledge]\n\n## Bundled Resources\n\n[Table of scripts, references, and assets with their purposes]\n```\n\n### Metadata Quality\n\nThe `name` and `description` in the frontmatter determine when the agent will load the skill. Write them carefully:\n\n- **name**: Use lowercase kebab-case (`my-skill`, not `MySkill` or `my_skill`)\n- **description**: Be specific about what the skill does and when to use it. Use third-person form (e.g., \"This skill should be used when...\" instead of \"Use this skill when...\")\n\n### Progressive Disclosure\n\nSkills use a three-level loading system to manage context efficiently:\n\n1. **Metadata** (name + description) — Always in the agent's system prompt (~100 words)\n2. **SKILL.md body** — Loaded when the agent calls `LoadSkill` (target: \u003C5k words)\n3. **Bundled resources** — Loaded on demand as the agent reads specific files (unlimited)\n\nKeep `SKILL.md` lean by moving detailed reference material to `references\u002F` files. Only include essential procedural instructions and workflow guidance in the main body.\n\n### Writing Style\n\n- Use **imperative\u002Finfinitive form** (verb-first instructions), not second person\n- Write \"To accomplish X, do Y\" rather than \"You should do X\"\n- Keep instructions actionable and specific — vague steps produce vague behavior\n- Reference bundled resources explicitly so the agent knows they exist and when to use them\n\n### Bundled Resources Guidelines\n\n| Type | Directory | When to Include | Example |\n|------|-----------|-----------------|---------|\n| Scripts | `scripts\u002F` | Same code is rewritten repeatedly or deterministic reliability is needed | `scripts\u002Fvalidate_config.py` |\n| References | `references\u002F` | Documentation the agent should consult while working | `references\u002Fapi_schema.md` |\n| Assets | `assets\u002F` | Files used in the output (templates, images, boilerplate) | `assets\u002Ftemplates\u002Fstarter.yaml` |\n\n**Avoid duplication**: Information should live in either `SKILL.md` or reference files, not both. For large reference files (>10k words), include grep search patterns in `SKILL.md` to help the agent locate relevant sections.\n\n### Choosing Between Skill Types\n\n| Criteria | Folder-Based | Tool-Based |\n|----------|-------------|------------|\n| Needs extensive documentation | ✓ | |\n| Includes multiple files (scripts, templates) | ✓ | |\n| Single, well-defined capability | | ✓ |\n| Tool description is sufficient documentation | | ✓ |\n| Will be shared across projects | ✓ | |\n\n### Common Pitfalls\n\n- Putting too much detail in `SKILL.md` instead of `references\u002F` — bloats context on every load\n- Vague frontmatter description — the agent cannot determine when to use the skill\n- Missing resource references — the agent does not know bundled files exist unless told\n- Absolute paths in documentation — use relative paths within the skill folder\n- Duplicating information between `SKILL.md` and reference files\n","# NexAU Agent Builder\n\nThis skill provides the procedural knowledge and reference material needed to implement NexAU agents correctly and efficiently.\n\n## When to Use\n\n- Implementing a new standalone agent (YAML config + system prompt + tools + entry point)\n- Adding custom tools to an existing agent\n- Writing or refining system prompts for NexAU agents\n- Setting up skills for an agent\n- Looking up NexAU framework concepts (agents, tools, LLMs, transports, sessions, etc.)\n\n> **NexAU Documentation**: The `docs\u002F` directory contains the full user-facing documentation for NexAU. Start with `docs\u002Findex.md` for an overview, `docs\u002Fgetting-started.md` for setup, `docs\u002Fcore-concepts\u002F` for fundamentals (agents, tools, LLMs), and `docs\u002Fadvanced-guides\u002F` for topics like skills, hooks, MCP, sandbox, transports, and session management.\n\n## Agent Implementation Workflow\n\n### 1. Understand the Requirements\n\nBefore writing any code, read the RFC or requirements document thoroughly. Identify:\n\n- Agent purpose and capabilities\n- Required tools (builtin vs custom)\n- LLM configuration needs (model, temperature, token limits)\n- Any skills the agent should have\n\n### 2. Create the Directory Structure\n\nA standalone agent follows this layout:\n\n```\nagent_name\u002F\n├── agent_name.yaml          # Agent configuration\n├── systemprompt.md           # System prompt (Jinja2 or plain string)\n├── start.py                  # Entry point script\n├── tools\u002F                    # Tool definitions\n│   ├── read_file.tool.yaml\n│   ├── write_file.tool.yaml\n│   └── custom_tool.tool.yaml\n├── skills\u002F                   # Optional skills\n│   └── skill-name\u002F\n│       └── SKILL.md\n└── custom_tools\u002F             # Custom tool implementations (if any)\n    └── custom_tool.py\n```\n\n\n### 3. Write the Agent YAML Config\n\nRefer to `references\u002Fagent-yaml-reference.md` for the complete field reference. Key points:\n\n- Use `${env.VAR_NAME}` for environment variable substitution (LLM keys, sandbox paths)\n- Set `system_prompt_type: jinja` when using template variables like `{{ date }}`\n- Set `system_prompt_type: string` for plain text prompts\n- Choose `tool_call_mode: openai` for OpenAI-compatible APIs\n- Reference tool YAMLs with relative paths: `.\u002Ftools\u002Ftool.tool.yaml`\n- Reference system prompt with relative path: `.\u002Fsystemprompt.md`\n\nTemplate available at: `assets\u002Ftemplates\u002Fagent.yaml`\n\n### 4. Define Tools\n\nFor each tool the agent needs:\n\n1. Check if a builtin tool exists — see `references\u002Fbuiltin-tools-reference.md`\n2. If builtin, copy the tool YAML from an existing example and set the binding path\n3. If custom, create both a `.tool.yaml` definition and a Python implementation\n\nRefer to `references\u002Ftool-yaml-reference.md` for the tool YAML schema.\n\nTemplate available at: `assets\u002Ftemplates\u002Ftool.tool.yaml`\n\n### 5. Write the System Prompt\n\nFollow the conventions in `references\u002Fsystem-prompt-guide.md`:\n\n- Start with a clear role description\n- Define a numbered step-by-step workflow\n- Include guidelines and constraints\n- Add template variables at the bottom: `{{ date }}`, `{{ username }}`, `{{ working_directory }}`\n\nTemplate available at: `assets\u002Ftemplates\u002Fsystemprompt.md`\n\n### 6. Create the Entry Point\n\nFor standalone agents, use `Agent.from_yaml()`.\n\nTemplate available at: `assets\u002Ftemplates\u002Fstart.py` (standalone)\n\n### 7. Verify the Implementation\n\n- **Validate agent config**:\n  ```bash\n  python {THIS_SKILL_FOLDER}\u002Fscripts\u002Fvalidate_agent.py path\u002Fto\u002Fagent.yaml\n  ```\n  - Add `--recursive` to also validate sub-agent configs\n  - Add `--json` for machine-readable output\n- Check Python syntax: `python -c \"import ast; ast.parse(open('start.py').read())\"`\n- Check imports: `python -c \"from nexau import Agent\"`\n- Run the agent with a simple test query\n\n## Common Patterns\n\n### Environment Variable Substitution in YAML\n\n```yaml\nllm_config:\n  model: ${env.LLM_MODEL}\n  base_url: ${env.LLM_BASE_URL}\n  api_key: ${env.LLM_API_KEY}\n```\n\n### Adding Builtin Tools\n\n```yaml\ntools:\n  - name: read_file\n    yaml_path: .\u002Ftools\u002Fread_file.tool.yaml\n    binding: nexau.archs.tool.builtin.file_tools:read_file\n```\n\n### Adding Skills\n\n```yaml\nskills:\n  - .\u002Fskills\u002Fskill-name\n```\n\n### Stop Tools\n\nUse `stop_tools` to define tools that halt the agent loop when called:\n\n```yaml\nstop_tools: [complete_task]    # For task-based agents\nstop_tools: [ask_user]         # For interactive agents\n```\n\n### Tracers\n\n```yaml\ntracers:\n  - import: nexau.archs.tracer.adapters.in_memory:InMemoryTracer\n```\n\n## Builtin Tools Catalog\n\nAll builtin tool YAML configs are located under `builtin_tools\u002Ftools\u002F` in this skill folder. When adding a builtin tool to an agent, copy the YAML to your agent's `tools\u002F` directory and set the binding path accordingly.\n\n### File Tools\n\n| Tool | Binding | Config YAML |\n|------|---------|-------------|\n| read_file | `nexau.archs.tool.builtin.file_tools:read_file` | `builtin_tools\u002Ftools\u002Fread_file.tool.yaml` |\n| write_file | `nexau.archs.tool.builtin.file_tools:write_file` | `builtin_tools\u002Ftools\u002Fwrite_file.tool.yaml` |\n| replace | `nexau.archs.tool.builtin.file_tools:replace` | `builtin_tools\u002Ftools\u002Freplace.tool.yaml` |\n| search_file_content | `nexau.archs.tool.builtin.file_tools:search_file_content` | `builtin_tools\u002Ftools\u002Fsearch_file_content.tool.yaml` |\n| glob | `nexau.archs.tool.builtin.file_tools:glob` | `builtin_tools\u002Ftools\u002FGlob.tool.yaml` |\n| list_directory | `nexau.archs.tool.builtin.file_tools:list_directory` | `builtin_tools\u002Ftools\u002Flist_directory.tool.yaml` |\n\n\n### Web Tools\n\n| Tool | Binding | Config YAML |\n|------|---------|-------------|\n| google_web_search | `nexau.archs.tool.builtin.web_tools:google_web_search` | `builtin_tools\u002Ftools\u002FWebSearch.tool.yaml` |\n| web_fetch | `nexau.archs.tool.builtin.web_tools:web_fetch` | `builtin_tools\u002Ftools\u002FWebFetch.tool.yaml` |\n\n### Shell Tools\n\n| Tool | Binding | Config YAML |\n|------|---------|-------------|\n| run_shell_command (sync) | `nexau.archs.tool.builtin.shell_tools:run_shell_command` | `builtin_tools\u002Ftools\u002Frun_shell_command_sync.tool.yaml` |\n\n### Session Tools\n\n| Tool | Binding | Config YAML |\n|------|---------|-------------|\n| write_todos | `nexau.archs.tool.builtin.session_tools:write_todos` | `builtin_tools\u002Ftools\u002Fwrite_todos.tool.yaml` |\n| complete_task | `nexau.archs.tool.builtin.session_tools:complete_task` | `builtin_tools\u002Ftools\u002Fcomplete_task.tool.yaml` |\n| ask_user | `nexau.archs.tool.builtin.session_tools:ask_user` | `builtin_tools\u002Ftools\u002Fask_user.tool.yaml` |\n\n\n> For detailed parameter descriptions of each tool, see `references\u002Fbuiltin-tools-reference.md`.\n\n## Bundled Resources\n\n| Path | Purpose |\n|------|---------|\n| `references\u002Fagent-yaml-reference.md` | Complete agent YAML config field reference |\n| `references\u002Ftool-yaml-reference.md` | Tool YAML definition schema and examples |\n| `references\u002Fbuiltin-tools-reference.md` | Catalog of all builtin tools with binding paths |\n| `references\u002Fsystem-prompt-guide.md` | System prompt writing conventions and patterns |\n| `assets\u002Ftemplates\u002Fagent.yaml` | Starter agent YAML config template |\n| `assets\u002Ftemplates\u002Ftool.tool.yaml` | Starter tool YAML definition template |\n| `assets\u002Ftemplates\u002Fsystemprompt.md` | Starter system prompt template |\n| `assets\u002Ftemplates\u002Fstart.py` | Standalone agent entry point template |\n| `scripts\u002Fvalidate_agent.py` | Agent YAML validator (schema + file reference checks) |\n| `docs\u002Findex.md` | NexAU documentation index — entry point for all user-facing docs |\n| `docs\u002Fgetting-started.md` | Installation, setup, and first agent walkthrough |\n| `docs\u002Fcore-concepts\u002Fagents.md` | Agent architecture, lifecycle, and configuration |\n| `docs\u002Fcore-concepts\u002Ftools.md` | Tool system — definition, binding, execution |\n| `docs\u002Fcore-concepts\u002Fllms.md` | LLM provider configuration and aggregator usage |\n| `docs\u002Fadvanced-guides\u002Fskills.md` | Skill system — folder-based and tool-based skills |\n| `docs\u002Fadvanced-guides\u002Fhooks.md` | Hook system for lifecycle event handling |\n| `docs\u002Fadvanced-guides\u002Fmcp.md` | Model Context Protocol integration |\n| `docs\u002Fadvanced-guides\u002Fsandbox.md` | Sandbox configuration and isolation |\n| `docs\u002Fadvanced-guides\u002Fsession-management.md` | Session persistence and history management |\n| `docs\u002Fadvanced-guides\u002Fstreaming-events.md` | Streaming event types and handling |\n| `docs\u002Fadvanced-guides\u002Ftransports.md` | Transport layer — HTTP, stdio, WebSocket, gRPC |\n| `docs\u002Fadvanced-guides\u002Ftemplating.md` | Jinja2 templating in system prompts and configs |\n| `docs\u002Fadvanced-guides\u002Fcontext_compaction.md` | Context compaction middleware for long conversations |\n| `docs\u002Fadvanced-guides\u002Fglobal-storage.md` | Global storage for cross-turn agent state |\n| `docs\u002Fadvanced-guides\u002Ftracer.md` | Tracing and observability setup |\n| `docs\u002Fadvanced-guides\u002Fimage.md` | Image input handling |\n\n## Setting Up Skills for an Agent\n\nSkills are reusable capabilities that extend an agent's knowledge and workflows. NexAU supports two types: **folder-based skills** and **tool-based skills**. Both are automatically registered and discoverable via the `LoadSkill` tool at runtime.\n\n### Folder-Based Skills\n\nA folder-based skill is a self-contained directory with a `SKILL.md` file. To add one to an agent:\n\n1. Create the skill directory under the agent's `skills\u002F` folder:\n\n```\nskills\u002F\n└── my-skill\u002F\n    ├── SKILL.md              # Required — metadata + instructions\n    ├── scripts\u002F              # Optional — executable code\n    ├── references\u002F           # Optional — context docs loaded on demand\n    └── assets\u002F               # Optional — templates, images, output files\n```\n\n2. Reference the skill in the agent YAML:\n\n```yaml\nskills:\n  - .\u002Fskills\u002Fmy-skill\n```\n\nThe agent will see a brief description of the skill in its system prompt and can call `LoadSkill` to read the full `SKILL.md` content when needed.\n\n### Tool-Based Skills\n\nA tool-based skill is a regular tool marked with `as_skill: true` in its YAML definition. This is useful for simple, well-defined capabilities that do not need extensive documentation.\n\n```yaml\n# tools\u002Fgenerate_code.tool.yaml\ntype: tool\nname: generate_code\ndescription: Generates code based on specifications\nas_skill: true\nskill_description: Code generation skill for multiple programming languages\ninput_schema:\n  type: object\n  properties:\n    language:\n      type: string\n    specification:\n      type: string\n  required: [language, specification]\n```\n\nThe `skill_description` field is required when `as_skill: true`. It provides the brief text shown in the skill registry.\n\n### Combining Both Types\n\nBoth skill types can coexist in the same agent:\n\n```yaml\n# Agent YAML\nskills:\n  - .\u002Fskills\u002Fdata-analysis       # Folder-based skill\n  - .\u002Fskills\u002Freport-writing      # Folder-based skill\n\ntools:\n  - name: web_search\n    yaml_path: .\u002Ftools\u002Fweb_search.tool.yaml   # Tool-based skill (as_skill: true)\n    binding: nexau.archs.tool.builtin.web_tools:google_web_search\n```\n\n### How Skills Work at Runtime\n\n1. All skills are registered in the agent's `skill_registry` during initialization\n2. A `LoadSkill` tool is automatically added to the agent\n3. The agent sees brief descriptions of all available skills in its system prompt\n4. When the agent needs detailed information, it calls `LoadSkill` with the skill name\n5. The full `SKILL.md` content (and skill folder path) is returned to the agent\n\n## Writing High-Quality Skills\n\nA well-crafted skill transforms a general-purpose agent into a specialized one. Follow these principles to maximize effectiveness.\n\n### SKILL.md Structure\n\nEvery `SKILL.md` must start with YAML frontmatter containing `name` and `description`:\n\n```markdown\n---\nname: my-skill\ndescription: Brief description of what this skill does. This skill should be used when [trigger conditions].\n---\n\n# Skill Title\n\n[Purpose — what this skill provides, in 1-2 sentences]\n\n## When to Use\n\n[Concrete trigger conditions — when should the agent load this skill]\n\n## Workflow \u002F Instructions\n\n[Step-by-step procedural knowledge]\n\n## Bundled Resources\n\n[Table of scripts, references, and assets with their purposes]\n```\n\n### Metadata Quality\n\nThe `name` and `description` in the frontmatter determine when the agent will load the skill. Write them carefully:\n\n- **name**: Use lowercase kebab-case (`my-skill`, not `MySkill` or `my_skill`)\n- **description**: Be specific about what the skill does and when to use it. Use third-person form (e.g., \"This skill should be used when...\" instead of \"Use this skill when...\")\n\n### Progressive Disclosure\n\nSkills use a three-level loading system to manage context efficiently:\n\n1. **Metadata** (name + description) — Always in the agent's system prompt (~100 words)\n2. **SKILL.md body** — Loaded when the agent calls `LoadSkill` (target: \u003C5k words)\n3. **Bundled resources** — Loaded on demand as the agent reads specific files (unlimited)\n\nKeep `SKILL.md` lean by moving detailed reference material to `references\u002F` files. Only include essential procedural instructions and workflow guidance in the main body.\n\n### Writing Style\n\n- Use **imperative\u002Finfinitive form** (verb-first instructions), not second person\n- Write \"To accomplish X, do Y\" rather than \"You should do X\"\n- Keep instructions actionable and specific — vague steps produce vague behavior\n- Reference bundled resources explicitly so the agent knows they exist and when to use them\n\n### Bundled Resources Guidelines\n\n| Type | Directory | When to Include | Example |\n|------|-----------|-----------------|---------|\n| Scripts | `scripts\u002F` | Same code is rewritten repeatedly or deterministic reliability is needed | `scripts\u002Fvalidate_config.py` |\n| References | `references\u002F` | Documentation the agent should consult while working | `references\u002Fapi_schema.md` |\n| Assets | `assets\u002F` | Files used in the output (templates, images, boilerplate) | `assets\u002Ftemplates\u002Fstarter.yaml` |\n\n**Avoid duplication**: Information should live in either `SKILL.md` or reference files, not both. For large reference files (>10k words), include grep search patterns in `SKILL.md` to help the agent locate relevant sections.\n\n### Choosing Between Skill Types\n\n| Criteria | Folder-Based | Tool-Based |\n|----------|-------------|------------|\n| Needs extensive documentation | ✓ | |\n| Includes multiple files (scripts, templates) | ✓ | |\n| Single, well-defined capability | | ✓ |\n| Tool description is sufficient documentation | | ✓ |\n| Will be shared across projects | ✓ | |\n\n### Common Pitfalls\n\n- Putting too much detail in `SKILL.md` instead of `references\u002F` — bloats context on every load\n- Vague frontmatter description — the agent cannot determine when to use the skill\n- Missing resource references — the agent does not know bundled files exist unless told\n- Absolute paths in documentation — use relative paths within the skill folder\n- Duplicating information between `SKILL.md` and reference files\n",{"name":5,"description":7},"main","0fd3b7baa36ecbae8cec971cc782eb05481545a7","examples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent","https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU","https:\u002F\u002Fgithub.com\u002FCharmCheen","https:\u002F\u002Fcharmcheen.github.io\u002F","learningggg code","2026-03-11 11:26:22",{"source":8,"skillName":5,"commands":45,"prompt":55,"zipAvailable":56},[46,49,52],{"tool":47,"command":48},"npx","npx skills add CharmCheen\u002Fschedule_assistant_based_on_NexAU --skill nexau-agent",{"tool":50,"command":51},"bunx","bunx skills add CharmCheen\u002Fschedule_assistant_based_on_NexAU --skill nexau-agent",{"tool":53,"command":54},"pnpm","pnpm dlx skills add CharmCheen\u002Fschedule_assistant_based_on_NexAU --skill nexau-agent","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002FSKILL.md",true,[58,64,70,74,78,84,88,92,96,100,104,108,112,116,120,124,128,132,135,139,143,147,151,155,159,163,167,171,175,179,183,187,191,195,199,203,207,211,215,219,223,227,231],{"path":59,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":62,"sourceHtmlUrl":63},"assets\u002Ftemplates\u002Fagent.yaml","ASSET","资产文件",2098,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fassets\u002Ftemplates\u002Fagent.yaml",{"path":65,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":68,"sourceHtmlUrl":69},"assets\u002Ftemplates\u002Fstart.py","SCRIPT","脚本",1081,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fassets\u002Ftemplates\u002Fstart.py",{"path":71,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":72,"sourceHtmlUrl":73},"assets\u002Ftemplates\u002Fsystemprompt.md",429,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fassets\u002Ftemplates\u002Fsystemprompt.md",{"path":75,"resourceType":60,"resourceTypeDesc":61,"sizeBytes":76,"sourceHtmlUrl":77},"assets\u002Ftemplates\u002Ftool.tool.yaml",881,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fassets\u002Ftemplates\u002Ftool.tool.yaml",{"path":79,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":82,"sourceHtmlUrl":83},"builtin_tools\u002Ftools\u002Fask_user.tool.yaml","REFERENCE","参考文档",2682,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002Fask_user.tool.yaml",{"path":85,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":86,"sourceHtmlUrl":87},"builtin_tools\u002Ftools\u002FBackgroundTaskManage.tool.yaml",1813,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002FBackgroundTaskManage.tool.yaml",{"path":89,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":90,"sourceHtmlUrl":91},"builtin_tools\u002Ftools\u002Fcomplete_task.tool.yaml",985,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002Fcomplete_task.tool.yaml",{"path":93,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":94,"sourceHtmlUrl":95},"builtin_tools\u002Ftools\u002FGlob.tool.yaml",1110,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002FGlob.tool.yaml",{"path":97,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":98,"sourceHtmlUrl":99},"builtin_tools\u002Ftools\u002Flist_directory.tool.yaml",1133,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002Flist_directory.tool.yaml",{"path":101,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":102,"sourceHtmlUrl":103},"builtin_tools\u002Ftools\u002Fread_file.tool.yaml",1057,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002Fread_file.tool.yaml",{"path":105,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":106,"sourceHtmlUrl":107},"builtin_tools\u002Ftools\u002Fread_many_files.tool.yaml",3549,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002Fread_many_files.tool.yaml",{"path":109,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":110,"sourceHtmlUrl":111},"builtin_tools\u002Ftools\u002Fread_visual_file.tool.yaml",1720,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002Fread_visual_file.tool.yaml",{"path":113,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":114,"sourceHtmlUrl":115},"builtin_tools\u002Ftools\u002Freplace.tool.yaml",4091,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002Freplace.tool.yaml",{"path":117,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":118,"sourceHtmlUrl":119},"builtin_tools\u002Ftools\u002Frun_shell_command_sync.tool.yaml",1350,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002Frun_shell_command_sync.tool.yaml",{"path":121,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":122,"sourceHtmlUrl":123},"builtin_tools\u002Ftools\u002Frun_shell_command.tool.yaml",2019,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002Frun_shell_command.tool.yaml",{"path":125,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":126,"sourceHtmlUrl":127},"builtin_tools\u002Ftools\u002Fsave_memory.tool.yaml",1469,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002Fsave_memory.tool.yaml",{"path":129,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":130,"sourceHtmlUrl":131},"builtin_tools\u002Ftools\u002Fsearch_file_content.tool.yaml",1264,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002Fsearch_file_content.tool.yaml",{"path":133,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":126,"sourceHtmlUrl":134},"builtin_tools\u002Ftools\u002FWebFetch.tool.yaml","https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002FWebFetch.tool.yaml",{"path":136,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":137,"sourceHtmlUrl":138},"builtin_tools\u002Ftools\u002FWebSearch.tool.yaml",1053,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002FWebSearch.tool.yaml",{"path":140,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":141,"sourceHtmlUrl":142},"builtin_tools\u002Ftools\u002Fwrite_file.tool.yaml",778,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002Fwrite_file.tool.yaml",{"path":144,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":145,"sourceHtmlUrl":146},"builtin_tools\u002Ftools\u002Fwrite_todos.tool.yaml",5910,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fbuiltin_tools\u002Ftools\u002Fwrite_todos.tool.yaml",{"path":148,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":149,"sourceHtmlUrl":150},"docs\u002Fadvanced-guides\u002Fcontext_compaction.md",4193,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fadvanced-guides\u002Fcontext_compaction.md",{"path":152,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":153,"sourceHtmlUrl":154},"docs\u002Fadvanced-guides\u002Fglobal-storage.md",7884,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fadvanced-guides\u002Fglobal-storage.md",{"path":156,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":157,"sourceHtmlUrl":158},"docs\u002Fadvanced-guides\u002Fhooks.md",7189,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fadvanced-guides\u002Fhooks.md",{"path":160,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":161,"sourceHtmlUrl":162},"docs\u002Fadvanced-guides\u002Fimage.md",4245,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fadvanced-guides\u002Fimage.md",{"path":164,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":165,"sourceHtmlUrl":166},"docs\u002Fadvanced-guides\u002Fmcp.md",1827,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fadvanced-guides\u002Fmcp.md",{"path":168,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":169,"sourceHtmlUrl":170},"docs\u002Fadvanced-guides\u002Fsandbox.md",32788,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fadvanced-guides\u002Fsandbox.md",{"path":172,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":173,"sourceHtmlUrl":174},"docs\u002Fadvanced-guides\u002Fsession-management.md",8980,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fadvanced-guides\u002Fsession-management.md",{"path":176,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":177,"sourceHtmlUrl":178},"docs\u002Fadvanced-guides\u002Fskills.md",13161,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fadvanced-guides\u002Fskills.md",{"path":180,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":181,"sourceHtmlUrl":182},"docs\u002Fadvanced-guides\u002Fstreaming-events.md",11615,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fadvanced-guides\u002Fstreaming-events.md",{"path":184,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":185,"sourceHtmlUrl":186},"docs\u002Fadvanced-guides\u002Ftemplating.md",1351,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fadvanced-guides\u002Ftemplating.md",{"path":188,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":189,"sourceHtmlUrl":190},"docs\u002Fadvanced-guides\u002Ftracer.md",3473,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fadvanced-guides\u002Ftracer.md",{"path":192,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":193,"sourceHtmlUrl":194},"docs\u002Fadvanced-guides\u002Ftransports.md",12717,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fadvanced-guides\u002Ftransports.md",{"path":196,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":197,"sourceHtmlUrl":198},"docs\u002Fcore-concepts\u002Fagents.md",3484,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fcore-concepts\u002Fagents.md",{"path":200,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":201,"sourceHtmlUrl":202},"docs\u002Fcore-concepts\u002Fllms.md",1262,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fcore-concepts\u002Fllms.md",{"path":204,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":205,"sourceHtmlUrl":206},"docs\u002Fcore-concepts\u002Ftools.md",5602,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fcore-concepts\u002Ftools.md",{"path":208,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":209,"sourceHtmlUrl":210},"docs\u002Fgetting-started.md",5362,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Fgetting-started.md",{"path":212,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":213,"sourceHtmlUrl":214},"docs\u002Findex.md",2721,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fdocs\u002Findex.md",{"path":216,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":217,"sourceHtmlUrl":218},"references\u002Fagent-yaml-reference.md",4330,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Freferences\u002Fagent-yaml-reference.md",{"path":220,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":221,"sourceHtmlUrl":222},"references\u002Fbuiltin-tools-reference.md",5660,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Freferences\u002Fbuiltin-tools-reference.md",{"path":224,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":225,"sourceHtmlUrl":226},"references\u002Fsystem-prompt-guide.md",4018,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Freferences\u002Fsystem-prompt-guide.md",{"path":228,"resourceType":80,"resourceTypeDesc":81,"sizeBytes":229,"sourceHtmlUrl":230},"references\u002Ftool-yaml-reference.md",3912,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Freferences\u002Ftool-yaml-reference.md",{"path":232,"resourceType":66,"resourceTypeDesc":67,"sizeBytes":233,"sourceHtmlUrl":234},"scripts\u002Fvalidate_agent.py",20476,"https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Fnexau-agent\u002Fscripts\u002Fvalidate_agent.py",[236,247,276],{"skillNo":237,"name":238,"nameZh":6,"description":239,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":240,"categories":241,"keywords":243,"publishedAt":31},"01M6DGW2X11R52091VJDWDBBDH","template-skill","Replace with description of the skill and when Claude should use it.","https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Ftemplate-skill\u002FSKILL.md",[242],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[244,245,246],"template","template skill","use",{"skillNo":248,"name":249,"nameZh":6,"description":250,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":251,"categories":252,"keywords":261,"publishedAt":31},"01M6DGVT96Y60BR3DVNEYEAHVG","rfc-writer","Guide for writing RFC (Request for Comments) documents following NexAU project conventions. This skill should be used when users want to create a new RFC, update an existing RFC, or need guidance on RFC structure, formatting, and best practices.","https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fnexau_building_team\u002Fskills\u002Frfc\u002FSKILL.md",[253,256],{"categoryNo":254,"categoryName":255,"taxonomyType":21,"taxonomyTypeDesc":22},"01JBSEEDCAT000000000000018","写作与文案",{"categoryNo":257,"categoryName":258,"taxonomyType":259,"taxonomyTypeDesc":260},"01JBSEEDCAT000000000000036","内容创作者","PROFESSION","适用职业",[262,263,264,265,266,267,268,269,29,270,271,272,273,274,275],"rfc writer","RFC","writer","formatting","project","conventions","structure","guidance","best practices","update","create","project conventions","documents","writing",{"skillNo":277,"name":238,"nameZh":6,"description":239,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":278,"categories":279,"keywords":281,"publishedAt":31},"01M6DGVCATM7CTKS0YGS5VDDS3","https:\u002F\u002Fgithub.com\u002FCharmCheen\u002Fschedule_assistant_based_on_NexAU\u002Fblob\u002Fmain\u002Fexamples\u002Fcode_agent\u002Fskills\u002Ftemplate-skill\u002FSKILL.md",[280],{"categoryNo":19,"categoryName":20,"taxonomyType":21,"taxonomyTypeDesc":22},[244,245,246]]