[{"data":1,"prerenderedAt":-1},["ShallowReactive",2],{"project-1037":3},{"id":4,"name":5,"fullName":6,"owner":7,"repo":5,"description":8,"homepage":9,"htmlUrl":10,"language":11,"languages":10,"totalLinesOfCode":10,"stars":12,"forks":13,"watchers":14,"openIssues":15,"contributorsCount":16,"subscribersCount":16,"size":16,"stars1d":17,"stars7d":18,"stars30d":19,"stars90d":16,"forks30d":16,"starsTrendScore":20,"compositeScore":21,"rankGlobal":10,"rankLanguage":10,"license":22,"archived":23,"fork":23,"defaultBranch":24,"hasWiki":25,"hasPages":23,"topics":26,"createdAt":10,"pushedAt":10,"updatedAt":38,"readmeContent":39,"aiSummary":40,"trendingCount":16,"starSnapshotCount":16,"syncStatus":41,"lastSyncTime":42,"discoverSource":43},1037,"graphify","safishamsi\u002Fgraphify","safishamsi","AI coding assistant skill (Claude Code, Codex, OpenCode, Cursor, Gemini CLI, and more). Turn any folder of code, SQL schemas, R scripts, shell scripts, docs, papers, images, or videos into a queryable knowledge graph. App code + database schema + infrastructure in one graph.","https:\u002F\u002Fgraphifylabs.ai\u002F",null,"Python",65486,6634,232,139,0,727,6249,18727,3800,45,"MIT License",false,"v7",true,[27,28,29,30,31,32,33,34,35,36,37],"antigravity","claude-code","codex","gemini","graphrag","knowledge-graph","leiden","openclaw","rag","skills","tree-sitter","2026-06-12 02:00:22","\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fgraphifylabs.ai\">\u003Cimg src=\"https:\u002F\u002Fraw.githubusercontent.com\u002Fsafishamsi\u002Fgraphify\u002Fv4\u002Fdocs\u002Flogo-text.svg\" width=\"260\" height=\"64\" alt=\"Graphify\"\u002F>\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  🇺🇸 \u003Ca href=\"README.md\">English\u003C\u002Fa> | 🇨🇳 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.zh-CN.md\">简体中文\u003C\u002Fa> | 🇯🇵 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.ja-JP.md\">日本語\u003C\u002Fa> | 🇰🇷 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.ko-KR.md\">한국어\u003C\u002Fa> | 🇩🇪 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.de-DE.md\">Deutsch\u003C\u002Fa> | 🇫🇷 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.fr-FR.md\">Français\u003C\u002Fa> | 🇪🇸 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.es-ES.md\">Español\u003C\u002Fa> | 🇮🇳 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.hi-IN.md\">हिन्दी\u003C\u002Fa> | 🇧🇷 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.pt-BR.md\">Português\u003C\u002Fa> | 🇷🇺 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.ru-RU.md\">Русский\u003C\u002Fa> | 🇸🇦 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.ar-SA.md\">العربية\u003C\u002Fa> | 🇮🇹 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.it-IT.md\">Italiano\u003C\u002Fa> | 🇵🇱 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.pl-PL.md\">Polski\u003C\u002Fa> | 🇳🇱 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.nl-NL.md\">Nederlands\u003C\u002Fa> | 🇹🇷 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.tr-TR.md\">Türkçe\u003C\u002Fa> | 🇺🇦 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.uk-UA.md\">Українська\u003C\u002Fa> | 🇻🇳 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.vi-VN.md\">Tiếng Việt\u003C\u002Fa> | 🇮🇩 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.id-ID.md\">Bahasa Indonesia\u003C\u002Fa> | 🇸🇪 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.sv-SE.md\">Svenska\u003C\u002Fa> | 🇬🇷 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.el-GR.md\">Ελληνικά\u003C\u002Fa> | 🇷🇴 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.ro-RO.md\">Română\u003C\u002Fa> | 🇨🇿 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.cs-CZ.md\">Čeština\u003C\u002Fa> | 🇫🇮 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.fi-FI.md\">Suomi\u003C\u002Fa> | 🇩🇰 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.da-DK.md\">Dansk\u003C\u002Fa> | 🇳🇴 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.no-NO.md\">Norsk\u003C\u002Fa> | 🇭🇺 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.hu-HU.md\">Magyar\u003C\u002Fa> | 🇹🇭 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.th-TH.md\">ภาษาไทย\u003C\u002Fa> | 🇹🇼 \u003Ca href=\"docs\u002Ftranslations\u002FREADME.zh-TW.md\">繁體中文\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fsafishamsi.gumroad.com\u002Fl\u002Fqetvlo\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FBook-The%20Memory%20Layer-2ea44f?style=flat&logo=gitbook&logoColor=white\" alt=\"The Memory Layer\"\u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fsafishamsi\u002Fgraphify\u002Factions\u002Fworkflows\u002Fci.yml\">\u003Cimg src=\"https:\u002F\u002Fgithub.com\u002Fsafishamsi\u002Fgraphify\u002Factions\u002Fworkflows\u002Fci.yml\u002Fbadge.svg?branch=v7\" alt=\"CI\"\u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fpypi.org\u002Fproject\u002Fgraphifyy\u002F\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fpypi\u002Fv\u002Fgraphifyy\" alt=\"PyPI\"\u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fpepy.tech\u002Fproject\u002Fgraphifyy\">\u003Cimg src=\"https:\u002F\u002Fstatic.pepy.tech\u002Fbadge\u002Fgraphifyy\" alt=\"Downloads\"\u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fgithub.com\u002Fsponsors\u002Fsafishamsi\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002Fsponsor-safishamsi-ea4aaa?logo=github-sponsors\" alt=\"Sponsor\"\u002F>\u003C\u002Fa>\n  \u003Ca href=\"https:\u002F\u002Fwww.linkedin.com\u002Fin\u002Fsafi-shamsi\">\u003Cimg src=\"https:\u002F\u002Fimg.shields.io\u002Fbadge\u002FLinkedIn-Safi%20Shamsi-0077B5?logo=linkedin\" alt=\"LinkedIn\"\u002F>\u003C\u002Fa>\n\u003C\u002Fp>\n\n\u003Cp align=\"center\">\n  \u003Ca href=\"https:\u002F\u002Fstar-history.com\u002F#safishamsi\u002Fgraphify&Date\">\n    \u003Cimg src=\"https:\u002F\u002Fapi.star-history.com\u002Fsvg?repos=safishamsi\u002Fgraphify&type=Date\" alt=\"Star History Chart\" width=\"600\"\u002F>\n  \u003C\u002Fa>\n\u003C\u002Fp>\n\nType `\u002Fgraphify` in your AI coding assistant and it maps your entire project — code, docs, PDFs, images, videos — into a knowledge graph you can query instead of grepping through files.\n\nWorks in Claude Code, Codex, OpenCode, Cursor, Gemini CLI, GitHub Copilot CLI, VS Code Copilot Chat, Aider, OpenClaw, Factory Droid, Trae, Hermes, Kiro, Pi, and Google Antigravity.\n\n```\n\u002Fgraphify .\n```\n\nThat's it. You get three files:\n\n```\ngraphify-out\u002F\n├── graph.html       open in any browser — click nodes, filter, search\n├── GRAPH_REPORT.md  the highlights: key concepts, surprising connections, suggested questions\n└── graph.json       the full graph — query it anytime without re-reading your files\n```\n\n---\n\n## Install\n\n**Requires Python 3.10+**\n\n```bash\nuv tool install graphifyy && graphify install\n# or: pipx install graphifyy && graphify install\n# or: pip install graphifyy && graphify install\n```\n\n> **Official package:** The PyPI package is `graphifyy` (double-y). Other `graphify*` packages on PyPI are not affiliated. The CLI command is still `graphify`.\n\n> **`graphify: command not found`?** Use `uv tool install graphifyy` or `pipx install graphifyy` — both put the CLI on PATH automatically. With plain `pip`, add `~\u002F.local\u002Fbin` (Linux) or `~\u002FLibrary\u002FPython\u002F3.x\u002Fbin` (Mac) to your PATH, or run `python -m graphify`.\n\n### Pick your platform\n\n| Platform | Install command |\n|----------|----------------|\n| Claude Code (Linux\u002FMac) | `graphify install` |\n| Claude Code (Windows) | `graphify install --platform windows` |\n| Codex | `graphify install --platform codex` |\n| OpenCode | `graphify install --platform opencode` |\n| GitHub Copilot CLI | `graphify install --platform copilot` |\n| VS Code Copilot Chat | `graphify vscode install` |\n| Aider | `graphify install --platform aider` |\n| OpenClaw | `graphify install --platform claw` |\n| Factory Droid | `graphify install --platform droid` |\n| Trae | `graphify install --platform trae` |\n| Trae CN | `graphify install --platform trae-cn` |\n| Gemini CLI | `graphify install --platform gemini` |\n| Hermes | `graphify install --platform hermes` |\n| Kiro IDE\u002FCLI | `graphify kiro install` |\n| Pi coding agent | `graphify install --platform pi` |\n| Cursor | `graphify cursor install` |\n| Google Antigravity | `graphify antigravity install` |\n\n> Codex users: also add `multi_agent = true` under `[features]` in `~\u002F.codex\u002Fconfig.toml`.\n> Codex uses `$graphify` instead of `\u002Fgraphify`.\n\n---\n\n## Make your assistant always use the graph\n\nRun this once in your project after building a graph:\n\n| Platform | Command |\n|----------|---------|\n| Claude Code | `graphify claude install` |\n| Codex | `graphify codex install` |\n| OpenCode | `graphify opencode install` |\n| GitHub Copilot CLI | `graphify copilot install` |\n| VS Code Copilot Chat | `graphify vscode install` |\n| Aider | `graphify aider install` |\n| OpenClaw | `graphify claw install` |\n| Factory Droid | `graphify droid install` |\n| Trae | `graphify trae install` |\n| Trae CN | `graphify trae-cn install` |\n| Cursor | `graphify cursor install` |\n| Gemini CLI | `graphify gemini install` |\n| Hermes | `graphify hermes install` |\n| Kiro IDE\u002FCLI | `graphify kiro install` |\n| Pi coding agent | `graphify pi install` |\n| Google Antigravity | `graphify antigravity install` |\n\nThis writes a small config file that tells your assistant to read `GRAPH_REPORT.md` before answering questions about your codebase. On platforms that support hooks (Claude Code, Codex, Gemini CLI), a hook fires automatically before every file-read call — your assistant navigates by the graph instead of grepping through everything.\n\nUninstall with the matching command (e.g. `graphify claude uninstall`).\n\n---\n\n## What's in the report\n\n- **God nodes** — the most-connected concepts in your project. Everything flows through these.\n- **Surprising connections** — links between things that live in different files or modules. Ranked by how unexpected they are.\n- **The \"why\"** — inline comments (`# NOTE:`, `# WHY:`, `# HACK:`), docstrings, and design rationale from docs are extracted as separate nodes linked to the code they explain.\n- **Suggested questions** — 4–5 questions the graph is uniquely positioned to answer.\n- **Confidence tags** — every inferred relationship is marked `EXTRACTED`, `INFERRED`, or `AMBIGUOUS`. You always know what was found vs guessed.\n\n---\n\n## What files it handles\n\n| Type | Extensions |\n|------|-----------|\n| Code (28 languages) | `.py .ts .js .jsx .tsx .go .rs .java .c .cpp .rb .cs .kt .scala .php .swift .lua .luau .zig .ps1 .ex .exs .m .jl .vue .svelte .groovy .gradle .sql .f .F .f90 .F90 .f95 .F95 .f03 .F03 .f08 .F08` |\n| Docs | `.md .mdx .html .txt .rst .yaml .yml` |\n| Office | `.docx .xlsx` (requires `pip install graphifyy[office]`) |\n| PDFs | `.pdf` |\n| Images | `.png .jpg .webp .gif` |\n| Video \u002F Audio | `.mp4 .mov .mp3 .wav` and more (requires `pip install graphifyy[video]`) |\n| YouTube \u002F URLs | any video URL (requires `pip install graphifyy[video]`) |\n\nCode is extracted locally with no API calls (AST via tree-sitter). Everything else goes through your AI assistant's model API.\n\n---\n\n## Common commands\n\n```bash\n\u002Fgraphify .                        # build graph for current folder\n\u002Fgraphify .\u002Fdocs --update          # re-extract only changed files\n\u002Fgraphify . --cluster-only         # rerun clustering without re-extracting\n\u002Fgraphify . --no-viz               # skip the HTML, just the report + JSON\n\u002Fgraphify . --wiki                 # build a markdown wiki from the graph\n\n\u002Fgraphify query \"what connects auth to the database?\"\n\u002Fgraphify path \"UserService\" \"DatabasePool\"\n\u002Fgraphify explain \"RateLimiter\"\n\n\u002Fgraphify add https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.03762   # fetch a paper and add it\n\u002Fgraphify add \u003Cyoutube-url>                       # transcribe and add a video\n\ngraphify hook install              # auto-rebuild on git commit\ngraphify merge-graphs a.json b.json              # combine two graphs\n```\n\nSee the [full command reference](#full-command-reference) below.\n\n---\n\n## Ignoring files\n\nCreate a `.graphifyignore` in your project root — same syntax as `.gitignore`, including `!` negation:\n\n```\n# .graphifyignore\nnode_modules\u002F\ndist\u002F\n*.generated.py\n\n# only index src\u002F, ignore everything else\n*\n!src\u002F\n!src\u002F**\n```\n\n---\n\n## Team setup\n\n`graphify-out\u002F` is meant to be committed to git so everyone on the team starts with a map.\n\n**Recommended `.gitignore` additions:**\n```\ngraphify-out\u002Fmanifest.json    # mtime-based, breaks after git clone\ngraphify-out\u002Fcost.json        # local only\n# graphify-out\u002Fcache\u002F         # optional: commit for speed, skip to keep repo small\n```\n\n**Workflow:**\n1. One person runs `\u002Fgraphify .` and commits `graphify-out\u002F`.\n2. Everyone pulls — their assistant reads the graph immediately.\n3. Run `graphify hook install` to auto-rebuild after each commit (AST only, no API cost). This also sets up a git merge driver so `graph.json` is never left with conflict markers — two devs committing in parallel get their graphs union-merged automatically.\n4. When docs or papers change, run `\u002Fgraphify --update` to refresh those nodes.\n\n---\n\n## Using the graph directly\n\n```bash\n# query the graph from the terminal\ngraphify query \"show the auth flow\"\ngraphify query \"what connects DigestAuth to Response?\" --graph graphify-out\u002Fgraph.json\n\n# expose the graph as an MCP server (for repeated tool-call access)\npython -m graphify.serve graphify-out\u002Fgraph.json\n```\n\nThe MCP server gives your assistant structured access: `query_graph`, `get_node`, `get_neighbors`, `shortest_path`.\n\n> **WSL \u002F Linux note:** Ubuntu ships `python3`, not `python`. Use a venv to avoid conflicts:\n> ```bash\n> python3 -m venv .venv && .venv\u002Fbin\u002Fpip install \"graphifyy[mcp]\"\n> ```\n\n---\n\n## Privacy\n\n- **Code files** — processed locally via tree-sitter. Nothing leaves your machine.\n- **Video \u002F audio** — transcribed locally with faster-whisper. Nothing leaves your machine.\n- **Docs, PDFs, images** — sent to your AI assistant for semantic extraction (via the `\u002Fgraphify` skill, using whatever model your IDE session runs). Headless `graphify extract` requires `GEMINI_API_KEY` \u002F `GOOGLE_API_KEY` (Gemini), `MOONSHOT_API_KEY` (Kimi), `ANTHROPIC_API_KEY` (Claude), `OPENAI_API_KEY` (OpenAI), or a running Ollama instance (`OLLAMA_BASE_URL`). The `--dedup-llm` flag uses the same key.\n- No telemetry, no usage tracking, no analytics.\n\n---\n\n## Full command reference\n\n```\n\u002Fgraphify                          # run on current directory\n\u002Fgraphify .\u002Fraw                    # run on a specific folder\n\u002Fgraphify .\u002Fraw --mode deep        # more aggressive relationship extraction\n\u002Fgraphify .\u002Fraw --update           # re-extract only changed files\n\u002Fgraphify .\u002Fraw --directed         # preserve edge direction\n\u002Fgraphify .\u002Fraw --cluster-only     # rerun clustering on existing graph\n\u002Fgraphify .\u002Fraw --no-viz           # skip HTML visualization\n\u002Fgraphify .\u002Fraw --obsidian         # generate Obsidian vault\n\u002Fgraphify .\u002Fraw --wiki             # build agent-crawlable markdown wiki\n\u002Fgraphify .\u002Fraw --svg              # export graph.svg\n\u002Fgraphify .\u002Fraw --graphml          # export for Gephi \u002F yEd\n\u002Fgraphify .\u002Fraw --neo4j            # generate cypher.txt for Neo4j\n\u002Fgraphify .\u002Fraw --neo4j-push bolt:\u002F\u002Flocalhost:7687\n\u002Fgraphify .\u002Fraw --watch            # auto-sync as files change\n\u002Fgraphify .\u002Fraw --mcp              # start MCP stdio server\n\n\u002Fgraphify add https:\u002F\u002Farxiv.org\u002Fabs\u002F1706.03762\n\u002Fgraphify add \u003Cvideo-url>\n\u002Fgraphify add https:\u002F\u002F... --author \"Name\" --contributor \"Name\"\n\n\u002Fgraphify query \"what connects attention to the optimizer?\"\n\u002Fgraphify query \"...\" --dfs --budget 1500\n\u002Fgraphify path \"DigestAuth\" \"Response\"\n\u002Fgraphify explain \"SwinTransformer\"\n\ngraphify hook install              # post-commit + post-checkout hooks\ngraphify hook uninstall\ngraphify hook status\n\ngraphify claude install \u002F uninstall\ngraphify codex install \u002F uninstall\ngraphify opencode install\ngraphify cursor install \u002F uninstall\ngraphify gemini install \u002F uninstall\ngraphify copilot install \u002F uninstall\ngraphify aider install \u002F uninstall\ngraphify claw install \u002F uninstall\ngraphify droid install \u002F uninstall\ngraphify trae install \u002F uninstall\ngraphify trae-cn install \u002F uninstall\ngraphify hermes install \u002F uninstall\ngraphify kiro install \u002F uninstall\ngraphify antigravity install \u002F uninstall\n\ngraphify extract .\u002Fdocs                        # headless LLM extraction for CI (no IDE needed)\ngraphify extract .\u002Fdocs --backend gemini       # explicit backend: gemini, kimi, claude, openai, or ollama\ngraphify extract .\u002Fdocs --backend gemini --model gemini-3.1-pro-preview\ngraphify extract .\u002Fdocs --backend ollama       # local Ollama (set OLLAMA_BASE_URL \u002F OLLAMA_MODEL)\ngraphify extract .\u002Fdocs --no-cluster           # raw extraction only, skip clustering\ngraphify extract .\u002Fdocs --dedup-llm            # LLM tiebreaker for ambiguous entity pairs (uses same API key)\ngraphify extract .\u002Fdocs --global --as myrepo   # extract and register into the cross-project global graph\nGRAPHIFY_MAX_OUTPUT_TOKENS=32768 graphify extract .\u002Fdocs --backend claude  # raise output cap for dense corpora\n\ngraphify global add graphify-out\u002Fgraph.json myrepo   # register a project graph into ~\u002F.graphify\u002Fglobal.json\ngraphify global remove myrepo                         # remove a project from the global graph\ngraphify global list                                  # show all registered repos + node\u002Fedge counts\ngraphify global path                                  # print path to the global graph file\n\ngraphify clone https:\u002F\u002Fgithub.com\u002Fkarpathy\u002FnanoGPT\ngraphify merge-graphs a.json b.json --out merged.json\ngraphify watch .\u002Fsrc\ngraphify check-update .\u002Fsrc\ngraphify update .\u002Fsrc\ngraphify cluster-only .\u002Fmy-project\ngraphify cluster-only .\u002Fmy-project --graph path\u002Fto\u002Fgraph.json  # custom graph location\n```\n\n---\n\n## Learn more\n\n- [How it works](docs\u002Fhow-it-works.md) — the extraction pipeline, community detection, confidence scoring, benchmarks\n- [ARCHITECTURE.md](ARCHITECTURE.md) — module breakdown, how to add a language\n- [Optional integrations](docs\u002Fdocker-mcp-sqlite.md) — Docker MCP Toolkit + SQLite\n\n---\n\n## Built on graphify — Penpax\n\n[**Penpax**](https:\u002F\u002Fgraphifylabs.ai) is the always-on layer built on top of graphify — it applies the same graph approach to your entire working life: meetings, browser history, emails, files, and code, updating continuously in the background.\n\nBuilt for people whose work lives across hundreds of conversations and documents they can never fully reconstruct. No cloud, fully on-device.\n\n**Free trial launching soon.** [Join the waitlist →](https:\u002F\u002Fgraphifylabs.ai)\n\n---\n\n\u003Cdetails>\n\u003Csummary>Contributing\u003C\u002Fsummary>\n\n**Worked examples** are the most useful contribution. Run `\u002Fgraphify` on a real corpus, save the output to `worked\u002F{slug}\u002F`, write an honest `review.md` covering what the graph got right and wrong, and open a PR.\n\n**Extraction bugs** — open an issue with the input file, the cache entry (`graphify-out\u002Fcache\u002F`), and what was missed or wrong.\n\nSee [ARCHITECTURE.md](ARCHITECTURE.md) for module responsibilities and how to add a language.\n\n\u003C\u002Fdetails>\n","Graphify 是一个AI编码助手工具，能够将代码文件夹、SQL模式、R脚本、Shell脚本、文档、论文、图片或视频转化为可查询的知识图谱。其核心功能包括通过先进的自然语言处理技术解析多种类型的文件，并将其结构化为统一的知识图谱，支持Claude Code, Codex, OpenCode, Cursor, Gemini CLI等多种AI编码技能。该工具特别适用于需要快速理解和导航复杂软件项目及其相关文档的场景，如大型软件开发、数据科学项目管理等。基于Python构建，遵循MIT许可证开源。",2,"2026-06-11 02:41:14","top_all"]