[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M7AHRGR47MW8GSRA4BSVAP8K":3,"skill-same:01M7AHRGR47MW8GSRA4BSVAP8K":111},{"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":18,"publishedAt":19,"linkOnly":20,"markdownContent":21,"markdownBody":22,"frontMatter":23,"refName":52,"commitSha":53,"skillDirectory":54,"forksCount":55,"repositoryHtmlUrl":56,"repositoryHomepageUrl":6,"ownerHtmlUrl":57,"ownerBlogUrl":58,"ownerDescription":59,"updatedAt":60,"install":61,"resources":74},"01M7AHRGR47MW8GSRA4BSVAP8K","code-puppy-agent",null,"How Code Puppy itself is built — its internal architecture, structure, codebase layout, and source modules. Explains agents, tools, the plugin\u002Fcallback hook system, models, MCP, sessions and history\u002Fcontext windows, skills (including skill namespaces for large catalogs), slash commands, config, messaging\u002FUI, system-prompt assembly, and i18n. Activate for ANY question about how Code Puppy works internally, why it behaves a certain way, where something lives in the code, how a feature is implemented, or how to navigate, debug, or extend the codebase (add a tool, agent, plugin, command, skill, model, or MCP server).","mpfaffenberger\u002Fcode_puppy_core_plugins","mpfaffenberger","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F5629756?v=4","01M7AHR64XS5N88RV7YHREA7M9","Mike Pfaffenberger","01M7AHR664JK4C9AVJM5NZ80YM",9,"MIT","https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fcode_puppy_agent\u002FSKILL.md",[],[],"2026-11-15 14:19:17",false,"---\nname: code-puppy-agent\ndescription: How Code Puppy itself is built — its internal architecture, structure, codebase layout, and source modules. Explains agents, tools, the plugin\u002Fcallback hook system, models, MCP, sessions and history\u002Fcontext windows, skills (including skill namespaces for large catalogs), slash commands, config, messaging\u002FUI, system-prompt assembly, and i18n. Activate for ANY question about how Code Puppy works internally, why it behaves a certain way, where something lives in the code, how a feature is implemented, or how to navigate, debug, or extend the codebase (add a tool, agent, plugin, command, skill, model, or MCP server).\nversion: \"2.0\"\nauthor: code-puppy\ntags:\n  - code-puppy\n  - architecture\n  - structure\n  - internals\n  - codebase\n  - source\n  - how-it-works\n  - self-awareness\n  - agents\n  - tools\n  - plugins\n  - hooks\n  - callbacks\n  - models\n  - mcp\n  - sessions\n  - history\n  - skills\n  - namespaces\n  - commands\n  - config\n  - messaging\n  - i18n\n  - extend\n  - debug\n  - reference\n---\n\n# Code Puppy — Architecture & Internals\n\nThis skill is a **self-awareness document**. When activated it teaches the\nLLM how Code Puppy is structured under the hood so it can navigate the\ncodebase, debug issues, and extend the product without guessing.\n\n> **Philosophy:** Code Puppy is plugin-first. Nearly all new functionality\n> should be a plugin in the separate `code-puppy-core-plugins` repo that hooks\n> into core via `code_puppy\u002Fcallbacks.py`. Don't edit\n> `code_puppy\u002Fcommand_line\u002F` or core agent files unless a hook genuinely\n> doesn't exist.\n\n> **Two repos.** Core (the `code-puppy` package) holds the runtime, agents,\n> tools, and the callback engine. Plugins (the `code-puppy-core-plugins`\n> package, this skill included) ship separately and are discovered through\n> `code_puppy.plugins` entry points. Paths below starting `code_puppy\u002F` are in\n> core; paths starting `code_puppy_core_plugins\u002F` are in the plugins repo.\n\nThis skill is split across multiple files so activating it doesn't force\nthe full ~30KB of internals into context on every use — this SKILL.md is\nthe index. **Read the reference file for the topic you actually need**;\neach one is self-contained and cites real file paths so you can jump\nstraight to source. All reference files are plain repo paths under\n`code_puppy_core_plugins\u002Fcode_puppy_agent\u002F` — read them with `read_file`.\n\n---\n\n## Layered Architecture\n\n```\n┌──────────────────────────────────────────────────┐\n│  TUI \u002F CLI  (command_line\u002F, messaging\u002F)           │\n│    user input → slash commands → agent dispatch   │\n├──────────────────────────────────────────────────┤\n│  Agent Layer  (agents\u002F)                           │\n│    BaseAgent → system prompt, tools, history      │\n│    _builder, _runtime, _compaction, _history      │\n├──────────────────────────────────────────────────┤\n│  Tool Layer  (tools\u002F)                             │\n│    TOOL_REGISTRY → register_tools_for_agent()     │\n├──────────────────────────────────────────────────┤\n│  Plugin Layer  (callbacks.py + plugins\u002F loader)   │\n│    register_callback(\"hook\", fn) at import time   │\n├──────────────────────────────────────────────────┤\n│  Model Layer  (model_factory.py, config.py)       │\n│    ModelFactory → pydantic-ai Model objects       │\n├──────────────────────────────────────────────────┤\n│  pydantic-ai  (external)                          │\n│    Agent.run(), streaming, tool schemas           │\n└──────────────────────────────────────────────────┘\n```\n\nThe TUI collects user input. It delegates to the **agent manager** which\nloads the current agent. The agent builds a pydantic-ai `Agent` with a\nsystem prompt + tool set, then `run_with_mcp()` streams the LLM response.\nPlugins hook into every stage via callbacks.\n\n---\n\n## Reference Map\n\n| File | Read this for |\n|------|----------------|\n| `AGENTS_AND_TOOLS.md` | How agents are structured\u002Fdiscovered (Python, JSON, plugin agents, precedence). How tools get registered and wired to an agent. Async-tool blocking-I\u002FO gotcha. |\n| `PLUGINS_AND_CALLBACKS.md` | Plugin discovery tiers, the callback hook table, in-place tool-result mutation and ordering, a minimal plugin example, and hook-system gotchas (`get_model_system_prompt` last-write-wins; don't do real work at import time). |\n| `MODELS_AND_MCP.md` | ModelFactory, model config precedence, model types. MCP server lifecycle, agent bindings, and a known gap (MCP servers can't yet provide skills). |\n| `SESSIONS_AND_HISTORY.md` | Message history, context-window compaction, session save\u002Fload commands. |\n| `SKILLS_SYSTEM.md` | Skill discovery\u002Factivation, bundled-resource-file gotcha for plugin-registered skills, and **skill namespaces** — how `namespace_skill_search` groups large skill catalogs and replaces the flat prompt list. |\n| `SYSTEM_PROMPT_CONFIG_AND_I18N.md` | Exact system-prompt assembly order, `puppy.cfg` + key directories, spill configuration, the messaging bus, and an i18n quick-reference (full guide: `docs\u002FI18N.md`). |\n\nDon't guess which file has what you need — the table above is exhaustive\nfor this skill's scope. If a question doesn't map cleanly to one row, it's\nprobably answered by combining two (e.g. \"why did my plugin's prompt\naddition get silently dropped\" is a `PLUGINS_AND_CALLBACKS.md` +\n`SYSTEM_PROMPT_CONFIG_AND_I18N.md` question).\n\n---\n\n## Development Conventions\n\n### Golden rules\n\n1. **Plugins over core** — if a callback hook exists, use it. Don't edit\n   `command_line\u002F` or core agent files.\n2. **One `register_callbacks.py` per plugin** — register at module scope.\n3. **600-line hard cap** per file — split into submodules. (This very\n   skill was split for exactly this reason — it was 632 lines as one file.)\n4. **Fail gracefully** — plugins must never crash the app. Wrap external\n   calls in try\u002Fexcept.\n5. **Return `None`** from hooks\u002Fcommands you don't own.\n6. **Always run linters** — `ruff check --fix`, `ruff format .`\n7. **Wrap user-facing strings** in `t()`\u002F`ngettext` (see\n   `SYSTEM_PROMPT_CONFIG_AND_I18N.md`) — don't hardcode English display\n   text. Model-facing system prompt content is explicitly exempt.\n8. **Never allow a Claude co-author commit.**\n\n### Plugin structure\n\n```\nmy_plugin\u002F\n├── __init__.py              # docstring (can be minimal)\n├── register_callbacks.py    # entry point: register_callback() calls\n├── helpers.py               # optional: logic split out\n└── README.md                # optional: documentation\n```\n\n### Zen of Code Puppy\n\n- Simple is better than complex.\n- Flat is better than nested.\n- If a hook exists for it, use it.\n- Files should be readable in one sitting.\n- The plugin system is the API surface; the core is the engine.\n\n---\n\n## Quick Reference: Key File Map\n\n| File | Responsibility |\n|------|---------------|\n| `agents\u002Fbase_agent.py` | Abstract agent base — thin conductor |\n| `agents\u002F_builder.py` | Builds pydantic-ai Agent + MCP wiring |\n| `agents\u002F_runtime.py` | `run_with_mcp()` — streaming, retries, cancellation |\n| `agents\u002F_compaction.py` | Context summarization |\n| `agents\u002Fagent_manager.py` | Agent registry, switching, discovery |\n| `agents\u002Fjson_agent.py` | JSON-config agent loader |\n| `tools\u002F__init__.py` | `TOOL_REGISTRY`, `register_tools_for_agent()` |\n| `plugins\u002F__init__.py` | Plugin loader (builtin → user → project) |\n| `callbacks.py` | Hook engine — 45+ phases |\n| `config.py` | Config read\u002Fwrite, directories, model settings |\n| `model_factory.py` | Model-name → pydantic-ai Model |\n| `mcp_\u002Fmanager.py` | MCP server lifecycle |\n| `session_storage.py` | Session pickle save\u002Fload |\n| `plugins\u002Fagent_skills\u002F` | Skills discovery, activation, UI |\n| `plugins\u002Fnamespace_skill_search\u002F` | Namespace grouping + on-demand browse tool for large skill catalogs |\n| `i18n\u002F` | Internationalization: catalogs, `t()`\u002F`ngettext`, locale detection, formatting |\n| `i18n\u002Flocales\u002F*.json` | Message catalogs (source of truth: `en-US.json`) |\n| `pydantic_patches.py` | Startup monkey-patches (clipboard fix, etc.) |\n\n---\n\n## When to use this skill\n\nActivate this skill when you need to:\n- Understand **how a feature works internally** before modifying it\n- **Create a new plugin** and need to know which hooks to use\n- **Debug** an agent, tool, model, skill, or MCP issue\n- **Navigate the codebase** and need to find the right file\n- **Extend Code Puppy** with custom tools, agents, skills, or commands\n\nThen read the specific reference file(s) from the map above — don't try\nto hold all six in context at once unless the question is genuinely\nthat broad.\n","# Code Puppy — Architecture & Internals\n\nThis skill is a **self-awareness document**. When activated it teaches the\nLLM how Code Puppy is structured under the hood so it can navigate the\ncodebase, debug issues, and extend the product without guessing.\n\n> **Philosophy:** Code Puppy is plugin-first. Nearly all new functionality\n> should be a plugin in the separate `code-puppy-core-plugins` repo that hooks\n> into core via `code_puppy\u002Fcallbacks.py`. Don't edit\n> `code_puppy\u002Fcommand_line\u002F` or core agent files unless a hook genuinely\n> doesn't exist.\n\n> **Two repos.** Core (the `code-puppy` package) holds the runtime, agents,\n> tools, and the callback engine. Plugins (the `code-puppy-core-plugins`\n> package, this skill included) ship separately and are discovered through\n> `code_puppy.plugins` entry points. Paths below starting `code_puppy\u002F` are in\n> core; paths starting `code_puppy_core_plugins\u002F` are in the plugins repo.\n\nThis skill is split across multiple files so activating it doesn't force\nthe full ~30KB of internals into context on every use — this SKILL.md is\nthe index. **Read the reference file for the topic you actually need**;\neach one is self-contained and cites real file paths so you can jump\nstraight to source. All reference files are plain repo paths under\n`code_puppy_core_plugins\u002Fcode_puppy_agent\u002F` — read them with `read_file`.\n\n---\n\n## Layered Architecture\n\n```\n┌──────────────────────────────────────────────────┐\n│  TUI \u002F CLI  (command_line\u002F, messaging\u002F)           │\n│    user input → slash commands → agent dispatch   │\n├──────────────────────────────────────────────────┤\n│  Agent Layer  (agents\u002F)                           │\n│    BaseAgent → system prompt, tools, history      │\n│    _builder, _runtime, _compaction, _history      │\n├──────────────────────────────────────────────────┤\n│  Tool Layer  (tools\u002F)                             │\n│    TOOL_REGISTRY → register_tools_for_agent()     │\n├──────────────────────────────────────────────────┤\n│  Plugin Layer  (callbacks.py + plugins\u002F loader)   │\n│    register_callback(\"hook\", fn) at import time   │\n├──────────────────────────────────────────────────┤\n│  Model Layer  (model_factory.py, config.py)       │\n│    ModelFactory → pydantic-ai Model objects       │\n├──────────────────────────────────────────────────┤\n│  pydantic-ai  (external)                          │\n│    Agent.run(), streaming, tool schemas           │\n└──────────────────────────────────────────────────┘\n```\n\nThe TUI collects user input. It delegates to the **agent manager** which\nloads the current agent. The agent builds a pydantic-ai `Agent` with a\nsystem prompt + tool set, then `run_with_mcp()` streams the LLM response.\nPlugins hook into every stage via callbacks.\n\n---\n\n## Reference Map\n\n| File | Read this for |\n|------|----------------|\n| `AGENTS_AND_TOOLS.md` | How agents are structured\u002Fdiscovered (Python, JSON, plugin agents, precedence). How tools get registered and wired to an agent. Async-tool blocking-I\u002FO gotcha. |\n| `PLUGINS_AND_CALLBACKS.md` | Plugin discovery tiers, the callback hook table, in-place tool-result mutation and ordering, a minimal plugin example, and hook-system gotchas (`get_model_system_prompt` last-write-wins; don't do real work at import time). |\n| `MODELS_AND_MCP.md` | ModelFactory, model config precedence, model types. MCP server lifecycle, agent bindings, and a known gap (MCP servers can't yet provide skills). |\n| `SESSIONS_AND_HISTORY.md` | Message history, context-window compaction, session save\u002Fload commands. |\n| `SKILLS_SYSTEM.md` | Skill discovery\u002Factivation, bundled-resource-file gotcha for plugin-registered skills, and **skill namespaces** — how `namespace_skill_search` groups large skill catalogs and replaces the flat prompt list. |\n| `SYSTEM_PROMPT_CONFIG_AND_I18N.md` | Exact system-prompt assembly order, `puppy.cfg` + key directories, spill configuration, the messaging bus, and an i18n quick-reference (full guide: `docs\u002FI18N.md`). |\n\nDon't guess which file has what you need — the table above is exhaustive\nfor this skill's scope. If a question doesn't map cleanly to one row, it's\nprobably answered by combining two (e.g. \"why did my plugin's prompt\naddition get silently dropped\" is a `PLUGINS_AND_CALLBACKS.md` +\n`SYSTEM_PROMPT_CONFIG_AND_I18N.md` question).\n\n---\n\n## Development Conventions\n\n### Golden rules\n\n1. **Plugins over core** — if a callback hook exists, use it. Don't edit\n   `command_line\u002F` or core agent files.\n2. **One `register_callbacks.py` per plugin** — register at module scope.\n3. **600-line hard cap** per file — split into submodules. (This very\n   skill was split for exactly this reason — it was 632 lines as one file.)\n4. **Fail gracefully** — plugins must never crash the app. Wrap external\n   calls in try\u002Fexcept.\n5. **Return `None`** from hooks\u002Fcommands you don't own.\n6. **Always run linters** — `ruff check --fix`, `ruff format .`\n7. **Wrap user-facing strings** in `t()`\u002F`ngettext` (see\n   `SYSTEM_PROMPT_CONFIG_AND_I18N.md`) — don't hardcode English display\n   text. Model-facing system prompt content is explicitly exempt.\n8. **Never allow a Claude co-author commit.**\n\n### Plugin structure\n\n```\nmy_plugin\u002F\n├── __init__.py              # docstring (can be minimal)\n├── register_callbacks.py    # entry point: register_callback() calls\n├── helpers.py               # optional: logic split out\n└── README.md                # optional: documentation\n```\n\n### Zen of Code Puppy\n\n- Simple is better than complex.\n- Flat is better than nested.\n- If a hook exists for it, use it.\n- Files should be readable in one sitting.\n- The plugin system is the API surface; the core is the engine.\n\n---\n\n## Quick Reference: Key File Map\n\n| File | Responsibility |\n|------|---------------|\n| `agents\u002Fbase_agent.py` | Abstract agent base — thin conductor |\n| `agents\u002F_builder.py` | Builds pydantic-ai Agent + MCP wiring |\n| `agents\u002F_runtime.py` | `run_with_mcp()` — streaming, retries, cancellation |\n| `agents\u002F_compaction.py` | Context summarization |\n| `agents\u002Fagent_manager.py` | Agent registry, switching, discovery |\n| `agents\u002Fjson_agent.py` | JSON-config agent loader |\n| `tools\u002F__init__.py` | `TOOL_REGISTRY`, `register_tools_for_agent()` |\n| `plugins\u002F__init__.py` | Plugin loader (builtin → user → project) |\n| `callbacks.py` | Hook engine — 45+ phases |\n| `config.py` | Config read\u002Fwrite, directories, model settings |\n| `model_factory.py` | Model-name → pydantic-ai Model |\n| `mcp_\u002Fmanager.py` | MCP server lifecycle |\n| `session_storage.py` | Session pickle save\u002Fload |\n| `plugins\u002Fagent_skills\u002F` | Skills discovery, activation, UI |\n| `plugins\u002Fnamespace_skill_search\u002F` | Namespace grouping + on-demand browse tool for large skill catalogs |\n| `i18n\u002F` | Internationalization: catalogs, `t()`\u002F`ngettext`, locale detection, formatting |\n| `i18n\u002Flocales\u002F*.json` | Message catalogs (source of truth: `en-US.json`) |\n| `pydantic_patches.py` | Startup monkey-patches (clipboard fix, etc.) |\n\n---\n\n## When to use this skill\n\nActivate this skill when you need to:\n- Understand **how a feature works internally** before modifying it\n- **Create a new plugin** and need to know which hooks to use\n- **Debug** an agent, tool, model, skill, or MCP issue\n- **Navigate the codebase** and need to find the right file\n- **Extend Code Puppy** with custom tools, agents, skills, or commands\n\nThen read the specific reference file(s) from the map above — don't try\nto hold all six in context at once unless the question is genuinely\nthat broad.\n",{"name":5,"tags":24,"author":25,"version":51,"description":7},[25,26,27,28,29,30,31,32,33,34,35,36,37,38,39,40,41,42,43,44,45,46,47,48,49,50],"code-puppy","architecture","structure","internals","codebase","source","how-it-works","self-awareness","agents","tools","plugins","hooks","callbacks","models","mcp","sessions","history","skills","namespaces","commands","config","messaging","i18n","extend","debug","reference","2.0","main","fb38cb76ff8cbc063f253a9dc90e5d34cd1fbf33","code_puppy_core_plugins\u002Fcode_puppy_agent",32,"https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins","https:\u002F\u002Fgithub.com\u002Fmpfaffenberger","https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy","Maintainer of Pydantic AI @pydantic. Creator of Code Puppy.","2026-11-11 23:56:01",{"source":8,"skillName":5,"commands":62,"prompt":72,"zipAvailable":73},[63,66,69],{"tool":64,"command":65},"npx","npx skills add mpfaffenberger\u002Fcode_puppy_core_plugins --skill code-puppy-agent",{"tool":67,"command":68},"bunx","bunx skills add mpfaffenberger\u002Fcode_puppy_core_plugins --skill code-puppy-agent",{"tool":70,"command":71},"pnpm","pnpm dlx skills add mpfaffenberger\u002Fcode_puppy_core_plugins --skill code-puppy-agent","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fcode_puppy_agent\u002FSKILL.md",true,[75,81,87,91,95,99,103,107],{"path":76,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":79,"sourceHtmlUrl":80},"__init__.py","SCRIPT","脚本",182,"https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fcode_puppy_agent\u002F__init__.py",{"path":82,"resourceType":83,"resourceTypeDesc":84,"sizeBytes":85,"sourceHtmlUrl":86},"AGENTS_AND_TOOLS.md","REFERENCE","参考文档",6406,"https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fcode_puppy_agent\u002FAGENTS_AND_TOOLS.md",{"path":88,"resourceType":83,"resourceTypeDesc":84,"sizeBytes":89,"sourceHtmlUrl":90},"MODELS_AND_MCP.md",3511,"https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fcode_puppy_agent\u002FMODELS_AND_MCP.md",{"path":92,"resourceType":83,"resourceTypeDesc":84,"sizeBytes":93,"sourceHtmlUrl":94},"PLUGINS_AND_CALLBACKS.md",8299,"https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fcode_puppy_agent\u002FPLUGINS_AND_CALLBACKS.md",{"path":96,"resourceType":77,"resourceTypeDesc":78,"sizeBytes":97,"sourceHtmlUrl":98},"register_callbacks.py",1916,"https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fcode_puppy_agent\u002Fregister_callbacks.py",{"path":100,"resourceType":83,"resourceTypeDesc":84,"sizeBytes":101,"sourceHtmlUrl":102},"SESSIONS_AND_HISTORY.md",1356,"https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fcode_puppy_agent\u002FSESSIONS_AND_HISTORY.md",{"path":104,"resourceType":83,"resourceTypeDesc":84,"sizeBytes":105,"sourceHtmlUrl":106},"SKILLS_SYSTEM.md",6416,"https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fcode_puppy_agent\u002FSKILLS_SYSTEM.md",{"path":108,"resourceType":83,"resourceTypeDesc":84,"sizeBytes":109,"sourceHtmlUrl":110},"SYSTEM_PROMPT_CONFIG_AND_I18N.md",8146,"https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fcode_puppy_agent\u002FSYSTEM_PROMPT_CONFIG_AND_I18N.md",[112,119,126,133,140],{"skillNo":113,"name":114,"nameZh":6,"description":115,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":116,"categories":117,"keywords":118,"publishedAt":19},"01M7AHS6Y4KY5ZDK4M25WAGY56","web-retriever","Use before handling web scraping, browser automation, crawling, structured data extraction, authenticated or interactive website workflows, monitoring pages for changes, or website screenshots. Delegates browser work to the web-retriever agent; not for test assertions or visual QA.","https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fweb_retriever_skill\u002FSKILL.md",[],[],{"skillNo":120,"name":121,"nameZh":6,"description":122,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":123,"categories":124,"keywords":125,"publishedAt":19},"01M7AHS2K4MKCX6648BT819T5Z","qa-kitten","Use before testing a web application's behavior, accessibility, responsive layout, or visual rendering in a real browser. Delegates browser assertions and visual QA to qa-kitten; not for scraping, crawling, or general web retrieval.","https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fqa_kitten_skill\u002FSKILL.md",[],[],{"skillNo":127,"name":128,"nameZh":6,"description":129,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":130,"categories":131,"keywords":132,"publishedAt":19},"01M7AHRXZAZPQWXPJV86KH0448","create-skill","Create, edit, test, or tune a Code Puppy skill (a SKILL.md with frontmatter and optional references\u002F). Use for \u002Fcreate-skill, \"make a skill for X\", \"turn this into a skill\", substantive edits to an existing skill, or \"tune description\" when a skill doesn't trigger when it should.","https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fpstack\u002Fskills\u002Fcreate-skill\u002FSKILL.md",[],[],{"skillNo":134,"name":135,"nameZh":6,"description":136,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":137,"categories":138,"keywords":139,"publishedAt":19},"01M7AHRRR11D2HBHYTCJE6K7Q7","herdr-code-puppy","Safely launch Code Puppy siblings and choose prompt versus explicit command input inside a Herdr pane. Activate before controlling siblings.","https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fherdr\u002FSKILL.md",[],[],{"skillNo":141,"name":142,"nameZh":6,"description":143,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":12,"repositoryNo":13,"starsCount":14,"licenseSpdxId":15,"licenseFileUrl":6,"sourceHtmlUrl":144,"categories":145,"keywords":146,"publishedAt":19},"01M7AHRA8TVNHZVRFQEG8TET4P","agent-creator","Use before creating or revising a Code Puppy JSON agent. Delegates schema design, least-privilege tool selection, optional model and MCP bindings, validation, and per-agent Spill configuration to agent-creator.","https:\u002F\u002Fgithub.com\u002Fmpfaffenberger\u002Fcode_puppy_core_plugins\u002Fblob\u002Fmain\u002Fcode_puppy_core_plugins\u002Fagent_creator_skill\u002FSKILL.md",[],[]]