Hugging Face MCP And Agent Spaces Enriched Dataset (77 features)
Last Updated: 06/01/2026 This dataset packages public Hugging Face Spaces discovered through MCP, agent, workflow, browser-use, and research search surfaces into a single analysis-ready dataframe. Each row represents one public Space enriched with Hub metadata, SDK and runtime details, search-surface provenance, repository-footprint signals, provider mentions, browser and workflow flags, MCP heuristics, and source URLs. The result is easier to use for agent-tool market mapping, competitive research, callable-app discovery, workflow monitoring, and ecosystem analysis than manually unioning Hub searches and inspecting Spaces one by one. Dataset includes 226 rows and 77 columns. Fields included record_id: Text column. Stable unique identifier for the dataset row derived from the Hugging Face Space id. space_id: Text column. Canonical Hugging Face Space identifier in owner/repo format. author: Text column. Owner slug for the Space, which may be a user or organization. repo_name: Text column. Repository name portion of the Space id. title: Text column. Human-readable Space title from the Hub card metadata when available. short_description: Text column. Short Hub description text for the Space. sdk: Text column. Primary Space SDK reported by the Hub, such as gradio, docker, or static. sdk_version: Text column. SDK version string from the Space card metadata when exposed. python_version: Text column. Python version declared in the Space card metadata when exposed. app_file: Text column. Primary application entry file declared in the Space card metadata when exposed. license_name: Text column. License string from the Space card metadata when exposed. card_tags: Text column. Pipe-delimited tags declared inside the Space card metadata. hub_tags: Text column. Pipe-delimited tags exposed by the Hub API for the Space. all_tags: Text column. Pipe-delimited union of card_tags and hub_tags after normalization and deduplication. card_tag_count: Whole-number numeric column. Number of tags declared inside the Space card metadata. hub_tag_count: Whole-number numeric column. Number of tags exposed by the Hub API. all_tag_count: Whole-number numeric column. Number of distinct tags after combining card and Hub tags. likes: Whole-number numeric column. Current Hub likes count for the public Space at collection time. trending_score: Decimal numeric column. Current Hub trending score reported by the Spaces API at collection time. created_at: Date or timestamp column. Timestamp when the Space repository was created on Hugging Face. last_modified_at: Date or timestamp column. Timestamp when the Space was last modified on Hugging Face. space_age_days: Whole-number numeric column. Number of days between Space creation and collection time. days_since_update: Whole-number numeric column. Number of days between last modification and collection time. is_recently_updated_30d: True/false column. Boolean flag indicating that the Space was updated within the last 30 days at collection time. private: True/false column. Boolean flag from the Hub detail API indicating whether the Space is private. gated: True/false column. Boolean flag indicating whether the Space is gated according to the Hub detail API. disabled: True/false column. Boolean flag indicating whether the Space is disabled according to the Hub detail API. runtime_stage: Column contains no non-null values. Runtime stage reported by the Hub detail API, such as RUNNING or BUILDING, when available. runtime_current_hardware: Column contains no non-null values. Current hardware tier reported by the Hub detail API when available. runtime_requested_hardware: Column contains no non-null values. Requested hardware tier reported by the Hub detail API when available. runtime_replica_count: Whole-number numeric column. Current replica count from the Hub detail API when available. runtime_dev_mode: True/false column. Boolean flag indicating whether the Space is currently in dev mode when exposed by the Hub detail API. subdomain: Text column. Hub-managed subdomain slug for the Space. host_url: Text column. Derived public hf.space host URL for the Space when available. space_page_url: Text column. Public Hugging Face page URL for the Space repository. sibling_count: Whole-number numeric column. Count of top-level sibling files exposed by the Hub detail API. python_file_count: Whole-number numeric column. Count of sibling files with Python extensions among the exposed top-level files. js_file_count: Whole-number numeric column. Count of sibling files with JavaScript or TypeScript extensions among the exposed top-level files. notebook_file_count: Whole-number numeric column. Count of sibling files with notebook extensions among the exposed top-level files. has_readme: True/false column. Boolean flag indicating that the top-level file listing includes a README. has_requirements_txt: True/false column. Boolean flag indicating that the top-level file listing includes requirements.txt. has_package_json: True/false column. Boolean flag indicating that the top-level file listing includes package.json. has_dockerfile: True/false column. Boolean flag indicating that the top-level file listing includes a Dockerfile. has_core_signal: True/false column. Boolean flag indicating that the Space text contains core MCP, agent, workflow, browser-use, automation, research, or assistant language. has_mcp_signal: True/false column. Boolean flag indicating explicit MCP or Model Context Protocol language. has_agent_signal: True/false column. Boolean flag indicating agent or agentic language. has_browser_use_signal: True/false column. Boolean flag indicating browser-use or browser-automation language. has_automation_signal: True/false column. Boolean flag indicating automation or orchestration language. has_research_signal: True/false column. Boolean flag indicating research or analysis language. has_workflow_signal: True/false column. Boolean flag indicating workflow or pipeline language. has_rag_signal: True/false column. Boolean flag indicating retrieval or RAG language. has_multimodal_signal: True/false column. Boolean flag indicating image, audio, speech, vision, or video language. has_tool_signal: True/false column. Boolean flag indicating tool, API, or function-calling language. has_chat_signal: True/false column. Boolean flag indicating chat or assistant language. mentions_openai: True/false column. Boolean flag indicating explicit OpenAI or ChatGPT mentions. mentions_anthropic: True/false column. Boolean flag indicating explicit Anthropic mentions. mentions_claude: True/false column. Boolean flag indicating explicit Claude mentions. mentions_gemini: True/false column. Boolean flag indicating explicit Gemini or Google AI mentions. mentions_deepseek: True/false column. Boolean flag indicating explicit DeepSeek mentions. mentions_ollama: True/false column. Boolean flag indicating explicit Ollama or local-model mentions. mentions_openrouter: True/false column. Boolean flag indicating explicit OpenRouter mentions. mentions_perplexity: True/false column. Boolean flag indicating explicit Perplexity mentions. provider_count: Whole-number numeric column. Count of distinct AI provider or assistant-brand mentions detected from the Space text and tags. matched_query_terms: Text column. Pipe-delimited list of search terms that surfaced the Space during collection. matched_query_term_count: Whole-number numeric column. Number of distinct search terms that surfaced the Space during collection. primary_query_term: Text column. First search term that surfaced the Space during collection. best_search_rank: Whole-number numeric column. Best rank position observed across the collected search surfaces. mean_search_rank: Decimal numeric column. Average rank position observed across the collected search surfaces. likely_gradio_api: True/false column. Boolean flag indicating that the Space uses the Gradio SDK and is likely callable through the Gradio API surface. likely_mcp_ready_space: True/false column. Boolean flag indicating that the Space has explicit MCP signals or an MCP-focused tag footprint. likely_browser_agent_space: True/false column. Boolean flag indicating that the Space combines agent or automation language with browser-use language. space_category: Text column. Primary product-positioning category inferred from the Space metadata and heuristics. source_api_list_url: Text column. Base Hugging Face Spaces list API used for search-driven collection. source_api_detail_url: Text column. Row-level Hugging Face Spaces detail API URL. source_directory_url: Text column. Public Hugging Face Spaces directory URL used as the source directory reference. source_domain: Text column. Domain name of the source Hub system used for this row. last_collected_at: Date or timestamp column. UTC timestamp when the row was collected and enriched. Find the full description and Dataset card with the dataset outline/structure/features on Hugging Face
Get it → thearticulated.gumroad.com