[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"skill:01M6BXX43ZBFPHTVTQAJV39KN5":3,"skill-same:01M6BXX43ZBFPHTVTQAJV39KN5":46},{"skillNo":4,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":15,"categories":16,"keywords":17,"publishedAt":20,"linkOnly":21,"markdownContent":22,"markdownBody":23,"frontMatter":24,"refName":26,"commitSha":27,"skillDirectory":28,"forksCount":13,"repositoryHtmlUrl":29,"repositoryHomepageUrl":6,"ownerHtmlUrl":30,"ownerBlogUrl":6,"ownerDescription":6,"updatedAt":31,"install":32,"resources":45},"01M6BXX43ZBFPHTVTQAJV39KN5","smart-tree",null,"Intelligent filesystem exploration using the `tree` command. Use this skill when you need to understand a project structure, find relevant files for a task, or map out a codebase before making changes. Replaces naive full-tree dumps with a multi-pass, heuristic-driven approach that minimizes token cost while maximizing structural insight.","NicholasGrundl\u002Fdockmaster","NicholasGrundl","https:\u002F\u002Favatars.githubusercontent.com\u002Fu\u002F46604911?v=4","01M6BXX1S9H9Q1XB8BWRTCYPW3","01M6BXX1Y11Q91EHK77DE012D4",0,"MIT","https:\u002F\u002Fgithub.com\u002FNicholasGrundl\u002Fdockmaster\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fsmart-tree\u002FSKILL.md",[],[18,19],"smart tree","project structure","2026-11-10 10:12:30",false,"---\nname: smart-tree\ndescription: >\n  Intelligent filesystem exploration using the `tree` command.\n  Use this skill when you need to understand a project structure, find relevant\n  files for a task, or map out a codebase before making changes. Replaces\n  naive full-tree dumps with a multi-pass, heuristic-driven approach that\n  minimizes token cost while maximizing structural insight.\nalwaysLoaded: false\n---\n\n# Smart Tree — Pure Scout\n\nYou are a **filesystem scout**. Your only job is to run `tree` commands, annotate\nthe output with importance markers, and return a structured map. The parent LLM\n(or user) decides what to read based on your map.\n\n**Hard rule:** While this skill is active, NEVER use file-reading tools (`read_file`,\n`grep`, `cat`, or any content-reading tool). You explore by **names and structure only**.\n\nUse your shell tool to execute `tree` commands.\n\n---\n\n## Annotation Vocabulary\n\nFive single-character symbols, right-aligned on each tree line, forming a visual\ngradient from signal to noise:\n\n```\n◆  key       — high value, read this first\n◇  useful    — worth reading, supporting context\n?  uncertain — might be relevant, needs human judgment\n·  noted     — acknowledged but low priority\n×  skip      — not relevant to the current goal, ignore\n```\n\nSymbols apply to both **files and directories**. A `◆` directory means \"high-value\ncontent inside.\" A `×` directory means \"don't bother descending.\"\n\nRight-side placement preserves tree indentation (the structural information) on\nthe left. Single-character width keeps annotations in a clean aligned column.\n\n---\n\n## Output Format\n\nEvery invocation produces a structured block with **five parts**:\n\n```\n## Smart Tree: \u002Fpath\u002Fto\u002Fproject\nGoal: understand the authentication system\nRecipe: find + deep\n\n├── src\u002F                     ◆\n│   ├── auth\u002F                ◆\n│   │   ├── middleware.py    ◆\n│   │   ├── tokens.py       ◆\n│   │   └── utils.py        ◇\n│   ├── api\u002F                 ◇\n│   │   ├── routes.py       ◇\n│   │   └── schemas.py      ·\n│   └── db\u002F                  ·\n│       ├── models.py        ◇\n│       └── migrations\u002F      ×\n├── tests\u002F                   ◇\n│   └── test_auth.py         ◆\n├── config\u002F                  ?\n│   ├── auth.yaml            ◆\n│   └── logging.yaml         ×\n├── node_modules\u002F            ×\n└── README.md                ◇\n\n◆ key · ◇ useful · ? uncertain · · noted · × skip\nAnnotations are based on filenames only, not content — treat ? items as worth a quick check.\n\n### Annotations\n- `auth\u002Fmiddleware.py` ◆ — likely contains the auth middleware referenced in the task\n- `auth\u002Ftokens.py` ◆ — JWT\u002Ftoken logic lives here based on name\n- `config\u002F` ? — could contain relevant settings, name alone doesn't confirm\n- `test_auth.py` ◆ — test coverage for the area we're changing\n\n### Recommended Read Order\n1. `src\u002Fauth\u002Fmiddleware.py` — entry point for auth flow\n2. `src\u002Fauth\u002Ftokens.py` — token logic\n3. `config\u002Fauth.yaml` — auth configuration\n4. `tests\u002Ftest_auth.py` — existing test coverage\n5. `src\u002Fapi\u002Froutes.py` — how auth integrates with API\n```\n\n**The five parts:**\n\n1. **Header** — path, goal, and recipe used (anchors context for the parent LLM)\n2. **Annotated tree** — the map itself; every visible item gets a symbol\n3. **Legend + calibration** — always present, always the same two lines\n4. **Annotation notes** — 3-5 bullets explaining `◆` and `?` choices (the reasoning)\n5. **Recommended Read Order** — numbered list of 3-7 files, ordered by priority\n\n---\n\n## Recipes\n\nThree compact recipes. Pick one or combine two. Max **3 tree calls** per invocation.\n\n| Recipe   | When                         | Command                                                         |\n|----------|------------------------------|-----------------------------------------------------------------|\n| **scan** | First look at a path         | `tree -L 2 --dirsfirst --noreport --gitignore`                  |\n| **find** | Looking for a specific topic | `tree -f --prune --ignore-case -P \"*topic*\" --gitignore`        |\n| **deep** | Drill into a known directory | `tree -L 3 --dirsfirst --noreport --gitignore \u003Cdir>`            |\n\nFor **find**, construct the `-P` pattern from your task context. Be creative with\nsynonyms: `\"*auth*|*login*|*session*|*oauth*|*jwt*\"`.\n\n---\n\n## Workflow\n\n1. **State intent** — Declare which recipe(s) you will use and why.\n   Example: *\"Recipe: scan + deep | Goal: map the agent runtime to find config files\"*\n\n2. **Execute** — Run 1-3 `tree` commands using your chosen recipe(s).\n   Use baseline flags on every call (see below). Present `tree` output verbatim.\n\n3. **Annotate** — Synthesize all `tree` outputs into a single annotated map.\n   Every visible item gets exactly one symbol. Combine outputs if you ran multiple commands.\n\n4. **Deliver** — Return the structured output (all five parts from the format above).\n\n---\n\n## Baseline Flags\n\nAlways include these on every `tree` call:\n\n```\n--dirsfirst     # dirs before files — easier to parse structure\n--noreport      # suppress summary line — saves tokens\n--gitignore     # respect .gitignore — auto-excludes build artifacts, deps, etc.\n```\n\nFor non-git directories or to exclude additional noise:\n\n```\n-I \"node_modules|.git|__pycache__|.venv|venv|.tox|.mypy_cache|.pytest_cache|.cache|.ruff_cache\"\n```\n\n---\n\n## Compact Flag Reference\n\n| Flag | Purpose | When to Use |\n|------|---------|-------------|\n| `-d` | Directories only | Skeleton scan — first pass of scan recipe |\n| `-L N` | Max depth N | Control recursion — start shallow (2), go deeper (3-4) |\n| `-f` | Print full path prefix | find recipe — need paths for later read calls |\n| `-P \"pat\"` | Only show matching files | find recipe — filter to specific names\u002Fextensions |\n| `--prune` | Remove empty dirs from filtered output | Always pair with `-P` |\n| `--filelimit=N` | Skip dirs with >N entries | Spot bloated dirs: 80 for wide scans, 50 for deep dives |\n\n> Full flag reference: `references\u002Ftree-manual.md`\n\n---\n\n## Patterns\n\nCommon intent-to-recipe mappings:\n\n**New repo \u002F unfamiliar codebase:**\n→ `scan` (skeleton overview) + `deep` into the source directory\nGoal: Build a mental map — language, framework, major subsystems.\n\n**Find code for a specific feature or topic:**\n→ `find` with topic synonyms + `deep` into matching directories\nGoal: Locate specific files to read. Pattern example: `\"*auth*|*login*|*session*\"`.\n\n**Plan where to add new code or docs:**\n→ `scan` (full layout) + `deep` into source dir AND docs\u002Fplanning dir\nGoal: Understand conventions — where do tests, configs, docs, and scripts live?\n\n**Investigate a bug or behavior:**\n→ `find` with error keywords or module names + `deep` into matches\nGoal: Narrow down to the 3-5 files most likely involved.\n\n---\n\n## Troubleshooting\n\nIf `tree` is not installed, inform the user and suggest installation:\n`apt install tree` (Debian\u002FUbuntu) · `brew install tree` (macOS) · `dnf install tree` (RHEL)\n","# Smart Tree — Pure Scout\n\nYou are a **filesystem scout**. Your only job is to run `tree` commands, annotate\nthe output with importance markers, and return a structured map. The parent LLM\n(or user) decides what to read based on your map.\n\n**Hard rule:** While this skill is active, NEVER use file-reading tools (`read_file`,\n`grep`, `cat`, or any content-reading tool). You explore by **names and structure only**.\n\nUse your shell tool to execute `tree` commands.\n\n---\n\n## Annotation Vocabulary\n\nFive single-character symbols, right-aligned on each tree line, forming a visual\ngradient from signal to noise:\n\n```\n◆  key       — high value, read this first\n◇  useful    — worth reading, supporting context\n?  uncertain — might be relevant, needs human judgment\n·  noted     — acknowledged but low priority\n×  skip      — not relevant to the current goal, ignore\n```\n\nSymbols apply to both **files and directories**. A `◆` directory means \"high-value\ncontent inside.\" A `×` directory means \"don't bother descending.\"\n\nRight-side placement preserves tree indentation (the structural information) on\nthe left. Single-character width keeps annotations in a clean aligned column.\n\n---\n\n## Output Format\n\nEvery invocation produces a structured block with **five parts**:\n\n```\n## Smart Tree: \u002Fpath\u002Fto\u002Fproject\nGoal: understand the authentication system\nRecipe: find + deep\n\n├── src\u002F                     ◆\n│   ├── auth\u002F                ◆\n│   │   ├── middleware.py    ◆\n│   │   ├── tokens.py       ◆\n│   │   └── utils.py        ◇\n│   ├── api\u002F                 ◇\n│   │   ├── routes.py       ◇\n│   │   └── schemas.py      ·\n│   └── db\u002F                  ·\n│       ├── models.py        ◇\n│       └── migrations\u002F      ×\n├── tests\u002F                   ◇\n│   └── test_auth.py         ◆\n├── config\u002F                  ?\n│   ├── auth.yaml            ◆\n│   └── logging.yaml         ×\n├── node_modules\u002F            ×\n└── README.md                ◇\n\n◆ key · ◇ useful · ? uncertain · · noted · × skip\nAnnotations are based on filenames only, not content — treat ? items as worth a quick check.\n\n### Annotations\n- `auth\u002Fmiddleware.py` ◆ — likely contains the auth middleware referenced in the task\n- `auth\u002Ftokens.py` ◆ — JWT\u002Ftoken logic lives here based on name\n- `config\u002F` ? — could contain relevant settings, name alone doesn't confirm\n- `test_auth.py` ◆ — test coverage for the area we're changing\n\n### Recommended Read Order\n1. `src\u002Fauth\u002Fmiddleware.py` — entry point for auth flow\n2. `src\u002Fauth\u002Ftokens.py` — token logic\n3. `config\u002Fauth.yaml` — auth configuration\n4. `tests\u002Ftest_auth.py` — existing test coverage\n5. `src\u002Fapi\u002Froutes.py` — how auth integrates with API\n```\n\n**The five parts:**\n\n1. **Header** — path, goal, and recipe used (anchors context for the parent LLM)\n2. **Annotated tree** — the map itself; every visible item gets a symbol\n3. **Legend + calibration** — always present, always the same two lines\n4. **Annotation notes** — 3-5 bullets explaining `◆` and `?` choices (the reasoning)\n5. **Recommended Read Order** — numbered list of 3-7 files, ordered by priority\n\n---\n\n## Recipes\n\nThree compact recipes. Pick one or combine two. Max **3 tree calls** per invocation.\n\n| Recipe   | When                         | Command                                                         |\n|----------|------------------------------|-----------------------------------------------------------------|\n| **scan** | First look at a path         | `tree -L 2 --dirsfirst --noreport --gitignore`                  |\n| **find** | Looking for a specific topic | `tree -f --prune --ignore-case -P \"*topic*\" --gitignore`        |\n| **deep** | Drill into a known directory | `tree -L 3 --dirsfirst --noreport --gitignore \u003Cdir>`            |\n\nFor **find**, construct the `-P` pattern from your task context. Be creative with\nsynonyms: `\"*auth*|*login*|*session*|*oauth*|*jwt*\"`.\n\n---\n\n## Workflow\n\n1. **State intent** — Declare which recipe(s) you will use and why.\n   Example: *\"Recipe: scan + deep | Goal: map the agent runtime to find config files\"*\n\n2. **Execute** — Run 1-3 `tree` commands using your chosen recipe(s).\n   Use baseline flags on every call (see below). Present `tree` output verbatim.\n\n3. **Annotate** — Synthesize all `tree` outputs into a single annotated map.\n   Every visible item gets exactly one symbol. Combine outputs if you ran multiple commands.\n\n4. **Deliver** — Return the structured output (all five parts from the format above).\n\n---\n\n## Baseline Flags\n\nAlways include these on every `tree` call:\n\n```\n--dirsfirst     # dirs before files — easier to parse structure\n--noreport      # suppress summary line — saves tokens\n--gitignore     # respect .gitignore — auto-excludes build artifacts, deps, etc.\n```\n\nFor non-git directories or to exclude additional noise:\n\n```\n-I \"node_modules|.git|__pycache__|.venv|venv|.tox|.mypy_cache|.pytest_cache|.cache|.ruff_cache\"\n```\n\n---\n\n## Compact Flag Reference\n\n| Flag | Purpose | When to Use |\n|------|---------|-------------|\n| `-d` | Directories only | Skeleton scan — first pass of scan recipe |\n| `-L N` | Max depth N | Control recursion — start shallow (2), go deeper (3-4) |\n| `-f` | Print full path prefix | find recipe — need paths for later read calls |\n| `-P \"pat\"` | Only show matching files | find recipe — filter to specific names\u002Fextensions |\n| `--prune` | Remove empty dirs from filtered output | Always pair with `-P` |\n| `--filelimit=N` | Skip dirs with >N entries | Spot bloated dirs: 80 for wide scans, 50 for deep dives |\n\n> Full flag reference: `references\u002Ftree-manual.md`\n\n---\n\n## Patterns\n\nCommon intent-to-recipe mappings:\n\n**New repo \u002F unfamiliar codebase:**\n→ `scan` (skeleton overview) + `deep` into the source directory\nGoal: Build a mental map — language, framework, major subsystems.\n\n**Find code for a specific feature or topic:**\n→ `find` with topic synonyms + `deep` into matching directories\nGoal: Locate specific files to read. Pattern example: `\"*auth*|*login*|*session*\"`.\n\n**Plan where to add new code or docs:**\n→ `scan` (full layout) + `deep` into source dir AND docs\u002Fplanning dir\nGoal: Understand conventions — where do tests, configs, docs, and scripts live?\n\n**Investigate a bug or behavior:**\n→ `find` with error keywords or module names + `deep` into matches\nGoal: Narrow down to the 3-5 files most likely involved.\n\n---\n\n## Troubleshooting\n\nIf `tree` is not installed, inform the user and suggest installation:\n`apt install tree` (Debian\u002FUbuntu) · `brew install tree` (macOS) · `dnf install tree` (RHEL)\n",{"name":5,"description":25,"alwaysLoaded":21},"Intelligent filesystem exploration using the `tree` command. Use this skill when you need to understand a project structure, find relevant files for a task, or map out a codebase before making changes. Replaces naive full-tree dumps with a multi-pass, heuristic-driven approach that minimizes token cost while maximizing structural insight.\n","main","96d23c843e14514104c3657dbe4b56f54a3b1126",".claude\u002Fskills\u002Fsmart-tree","https:\u002F\u002Fgithub.com\u002FNicholasGrundl\u002Fdockmaster","https:\u002F\u002Fgithub.com\u002FNicholasGrundl","2026-03-31 03:29:25",{"source":8,"skillName":5,"commands":33,"prompt":43,"zipAvailable":44},[34,37,40],{"tool":35,"command":36},"npx","npx skills add NicholasGrundl\u002Fdockmaster --skill smart-tree",{"tool":38,"command":39},"bunx","bunx skills add NicholasGrundl\u002Fdockmaster --skill smart-tree",{"tool":41,"command":42},"pnpm","pnpm dlx skills add NicholasGrundl\u002Fdockmaster --skill smart-tree","安装技能，技能地址：https:\u002F\u002Fgithub.com\u002FNicholasGrundl\u002Fdockmaster\u002Fblob\u002Fmain\u002F.claude\u002Fskills\u002Fsmart-tree\u002FSKILL.md",true,[],[47,52,76,92,98,108],{"skillNo":48,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":49,"categories":50,"keywords":51,"publishedAt":20},"01M6BXXKSNAMMW6M4F5JJV62W9","https:\u002F\u002Fgithub.com\u002FNicholasGrundl\u002Fdockmaster\u002Fblob\u002Fmain\u002F_blueprint\u002Fskills\u002Fsmart-tree\u002FSKILL.md",[],[18,19],{"skillNo":53,"name":54,"nameZh":6,"description":55,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":56,"categories":57,"keywords":63,"publishedAt":20},"01M6BXXGAZ8J8TDQSBVP662T5B","playwright-cli","Automates browser interactions for web testing, form filling, screenshots, and data extraction. Use when the user needs to navigate websites, interact with web pages, fill forms, take screenshots, test web applications, or extract information from web pages.","https:\u002F\u002Fgithub.com\u002FNicholasGrundl\u002Fdockmaster\u002Fblob\u002Fmain\u002F_blueprint\u002Fskills\u002Fplaywright-cli\u002FSKILL.md",[58],{"categoryNo":59,"categoryName":60,"taxonomyType":61,"taxonomyTypeDesc":62},"01JBSEEDCAT000000000000008","测试与质量","SCENARIO","使用场景",[64,65,66,67,68,69,70,71,72,73,74,75],"playwright cli","playwright","cli","browser","testing","screenshots","test","web","extraction","data","applications","Bash",{"skillNo":77,"name":78,"nameZh":6,"description":79,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":80,"categories":81,"keywords":85,"publishedAt":20},"01M6BXXDBQRV33XR4A58EG7WKN","context-adder","Workflow for adding new reference codebases (context) to the project's blueprint. Use this skill when you need to pull an external repository, summarize its structure, and integrate it into the project's documentation (SYLLABUS.md).","https:\u002F\u002Fgithub.com\u002FNicholasGrundl\u002Fdockmaster\u002Fblob\u002Fmain\u002F_blueprint\u002Fskills\u002Fcontext-adder\u002FSKILL.md",[82],{"categoryNo":83,"categoryName":84,"taxonomyType":61,"taxonomyTypeDesc":62},"01JBSEEDCAT000000000000023","Skill与智能体开发",[86,87,88,89,90,91],"context adder","repository","pull","documentation","workflow","summarize",{"skillNo":93,"name":54,"nameZh":6,"description":55,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":94,"categories":95,"keywords":97,"publishedAt":20},"01M6BXXA24EJ1DV1GWFDZRCDT1","https:\u002F\u002Fgithub.com\u002FNicholasGrundl\u002Fdockmaster\u002Fblob\u002Fmain\u002F_blueprint\u002Fcontext\u002Fplaywright-cli\u002Frepo\u002Fskills\u002Fplaywright-cli\u002FSKILL.md",[96],{"categoryNo":59,"categoryName":60,"taxonomyType":61,"taxonomyTypeDesc":62},[64,65,66,68,75,74,73,72,67,71,70,69],{"skillNo":99,"name":100,"nameZh":6,"description":101,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":102,"categories":103,"keywords":104,"publishedAt":20},"01M6BXX8677W6AW69CBQTHC4QF","dev","Development workflows for the playwright-cli repository. Use when the user asks about rolling dependencies, releasing, or other repo maintenance tasks.","https:\u002F\u002Fgithub.com\u002FNicholasGrundl\u002Fdockmaster\u002Fblob\u002Fmain\u002F_blueprint\u002Fcontext\u002Fplaywright-cli\u002Frepo\u002F.claude\u002Fskills\u002Fdev\u002FSKILL.md",[],[100,105,106,65,66,107,87],"tasks","workflows","development",{"skillNo":109,"name":5,"nameZh":6,"description":7,"descriptionZh":6,"repositoryFullName":8,"ownerLogin":9,"ownerAvatarUrl":10,"ownerAccountNo":11,"ownerDisplayName":6,"repositoryNo":12,"starsCount":13,"licenseSpdxId":14,"licenseFileUrl":6,"sourceHtmlUrl":110,"categories":111,"keywords":112,"publishedAt":20},"01M6BXX6NFDH9EJZN7WQMJ0TS1","https:\u002F\u002Fgithub.com\u002FNicholasGrundl\u002Fdockmaster\u002Fblob\u002Fmain\u002F.gemini\u002Fskills\u002Fsmart-tree\u002FSKILL.md",[],[18,19]]