{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"vscode-ms-windows-ai-studio-windows-ai-studio","slug":"foundry-toolkit-for-vs-code","name":"Foundry Toolkit for VS Code","type":"extension","url":"https://marketplace.visualstudio.com/items?itemName=ms-windows-ai-studio.windows-ai-studio","page_url":"https://unfragile.ai/foundry-toolkit-for-vs-code","categories":["app-builders"],"tags":["ai","Azure","Chat","Data Science","jsonl","language-models","language-model-tools","LLM","Machine Learning","models","python","studio","tools","windows"],"pricing":{"model":"freemium","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"vscode-ms-windows-ai-studio-windows-ai-studio__cap_0","uri":"capability://search.retrieval.multi.source.model.discovery.and.catalog.browsing","name":"multi-source model discovery and catalog browsing","description":"Provides a unified model discovery interface within VS Code that aggregates models from 8+ sources (Microsoft Foundry, GitHub Models, OpenAI, Anthropic, Google, NVIDIA NIM, Ollama, ONNX) with side-by-side comparison capabilities. The extension maintains a tree view in the sidebar with a 'Model Catalog' section that dynamically populates available models based on configured API keys and local installations, enabling developers to evaluate and select models without leaving the editor.","intents":["Compare capabilities and pricing across multiple LLM providers in one interface","Discover open-source models available locally via Ollama or ONNX","Find the right model for my use case by browsing curated catalogs","Quickly switch between proprietary and open models for experimentation"],"best_for":["AI/ML engineers evaluating multiple model providers","Solo developers prototyping with different LLM backends","Teams standardizing on model selection across projects"],"limitations":["Model catalog population requires valid API keys for proprietary providers (OpenAI, Anthropic, Google) or local installation (Ollama, ONNX)","No built-in model performance benchmarking — comparison is metadata-only (pricing, context window, capabilities)","Model availability depends on provider API status and network connectivity"],"requires":["VS Code (version unspecified, likely 1.80+)","Windows OS (implied by 'Windows AI Studio' branding)","API keys for proprietary models (optional for local models via Ollama/ONNX)"],"input_types":["configuration (API keys, local model paths)"],"output_types":["model metadata (name, provider, context window, pricing, capabilities)"],"categories":["search-retrieval","model-discovery"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-ms-windows-ai-studio-windows-ai-studio__cap_1","uri":"capability://text.generation.language.interactive.model.playground.with.multi.modal.input","name":"interactive model playground with multi-modal input","description":"Provides an embedded chat interface within VS Code for real-time model testing and prompt experimentation. The playground supports multi-modal inputs (text, images, attachments), parameter tuning (temperature, top-p, max tokens), and streaming response visualization. Developers can test prompts against any model in the catalog without leaving the editor, with full parameter control and response inspection.","intents":["Test and iterate on prompts before integrating into agent code","Experiment with different models and parameters to optimize quality","Debug model behavior with multi-modal inputs (text + images)","Validate model output format before building downstream workflows"],"best_for":["Prompt engineers iterating on system prompts and few-shot examples","Developers prototyping multi-modal AI features","Teams validating model behavior before production deployment"],"limitations":["Playground operates in-memory — no persistent conversation history or export functionality documented","Parameter tuning UI scope unknown (unclear which parameters are exposed for each model type)","Multi-modal input support limited to images and attachments (no video, audio, or custom formats documented)","No built-in cost tracking for API calls made during experimentation"],"requires":["VS Code with Foundry Toolkit extension installed","Selected model must be accessible (API key for cloud models, local installation for Ollama/ONNX)","Network connectivity for cloud-based models"],"input_types":["text (prompt)","image (JPEG, PNG, WebP — formats unspecified)","attachments (file types unspecified)"],"output_types":["text (streamed model response)","structured data (if model configured for JSON output)"],"categories":["text-generation-language","image-visual"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-ms-windows-ai-studio-windows-ai-studio__cap_10","uri":"capability://planning.reasoning.multi.model.agent.orchestration.and.comparison","name":"multi-model agent orchestration and comparison","description":"Enables agents to route requests to multiple models simultaneously or sequentially, compare outputs, and select the best response based on custom criteria. The extension provides orchestration patterns (parallel execution, fallback chains, ensemble voting) and comparison metrics (similarity, relevance, cost) to help developers optimize agent behavior. Results from all models are captured and compared in the debugger.","intents":["Compare model outputs for the same prompt to select the best response","Implement fallback chains where secondary models handle failures from primary models","Ensemble multiple models to improve response quality","Optimize cost by routing to cheaper models when quality is sufficient"],"best_for":["Teams optimizing model selection for quality and cost","Developers building resilient agents with fallback strategies","Organizations using ensemble methods for improved accuracy"],"limitations":["Supported orchestration patterns unknown (unclear if all patterns mentioned are implemented)","Model selection criteria and comparison metrics scope unknown","Latency impact of parallel execution and comparison not documented","Cost optimization logic and decision criteria unknown","No built-in statistical significance testing for model comparison","Ensemble voting mechanism and weighting options unknown"],"requires":["VS Code with Foundry Toolkit extension","Multiple models configured and accessible","Agent code with orchestration logic (built-in patterns or custom, unclear)"],"input_types":["agent prompt/input","model list (references to catalog models)","orchestration configuration (pattern selection, comparison criteria)"],"output_types":["model responses (from all models)","comparison results (metrics, selected response)","execution traces (per-model latency, cost)"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-ms-windows-ai-studio-windows-ai-studio__cap_11","uri":"capability://automation.workflow.agent.deployment.and.lifecycle.management","name":"agent deployment and lifecycle management","description":"Manages agent deployment to Microsoft Foundry and other hosting environments, including versioning, rollback, and environment configuration. Developers can deploy agents directly from VS Code, manage multiple versions, configure environment-specific settings (API keys, model selections), and monitor deployed agent health. The extension handles deployment packaging and orchestrates the deployment process.","intents":["Deploy agents from development to production without manual packaging","Manage multiple agent versions and rollback to previous versions if needed","Configure environment-specific settings (dev, staging, production)","Monitor deployed agent health and performance"],"best_for":["Teams deploying agents to production","Organizations managing multiple agent versions","Developers automating agent deployment workflows"],"limitations":["Supported deployment targets unknown (unclear if only Foundry or other platforms supported)","Deployment packaging format and requirements unknown","Environment configuration scope and validation unknown","Health monitoring metrics and alerting not documented","Rollback mechanism and version retention policy unknown","Deployment approval workflows and access control not documented"],"requires":["VS Code with Foundry Toolkit extension","Agent code in supported format","Microsoft Foundry account (for Foundry deployment)","Deployment credentials and permissions"],"input_types":["agent code (Python/JavaScript)","deployment configuration (environment, settings)","version metadata (description, tags)"],"output_types":["deployment package (format unspecified)","deployment status (success/failure, logs)","deployed agent URL/endpoint"],"categories":["automation-workflow","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-ms-windows-ai-studio-windows-ai-studio__cap_2","uri":"capability://code.generation.editing.no.code.and.code.based.agent.builder.with.structured.output","name":"no-code and code-based agent builder with structured output","description":"Provides dual-mode agent development: a no-code prompt-based agent builder for simple workflows and a code-based hosted agent framework for complex multi-step agents. Both modes support structured output generation (JSON schemas, typed responses) and integrate with the debugger for real-time execution visualization. The builder abstracts away boilerplate agent scaffolding while maintaining full code access for advanced customization.","intents":["Build simple agents with prompts and tools without writing orchestration code","Develop complex multi-step agents with conditional logic and state management","Generate structured outputs (JSON, typed objects) from agent responses","Debug agent execution flow and tool calls in real-time"],"best_for":["Non-technical users building simple chatbots or task agents","Full-stack developers building production multi-agent systems","Teams standardizing on agent patterns across projects"],"limitations":["No-code builder scope and supported agent patterns unknown (unclear what agent types are supported without code)","Code-based agents require Python or JavaScript runtime (language support unspecified)","Structured output validation and schema enforcement mechanism unknown","Agent state persistence and multi-turn conversation management not documented","No built-in agent versioning or rollback capabilities documented"],"requires":["VS Code with Foundry Toolkit extension","Python 3.9+ or Node.js 18+ (for code-based agents, versions unspecified)","Selected model accessible via API or local installation"],"input_types":["text (prompts, system instructions)","code (Python/JavaScript agent definitions)","tool definitions (function signatures, descriptions)"],"output_types":["agent code (Python/JavaScript)","structured responses (JSON, typed objects)","execution logs (for debugging)"],"categories":["code-generation-editing","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-ms-windows-ai-studio-windows-ai-studio__cap_3","uri":"capability://planning.reasoning.agent.execution.debugging.with.streaming.visualization","name":"agent execution debugging with streaming visualization","description":"Provides F5-based debugger integration for agent execution with real-time streaming response visualization and multi-agent workflow inspection. When launching an agent with F5, the extension captures execution traces, tool calls, and model responses, displaying them in a structured timeline view within VS Code. Developers can inspect intermediate states, tool invocations, and response generation without external logging or debugging tools.","intents":["Debug agent execution flow and identify where agents fail or behave unexpectedly","Inspect tool calls and their outputs in real-time during agent execution","Visualize multi-agent interactions and communication patterns","Trace model response generation and token streaming"],"best_for":["AI engineers debugging complex agent workflows","Teams troubleshooting production agent issues","Developers optimizing agent performance and latency"],"limitations":["Debugger integration scope unclear (unclear which agent frameworks are supported)","Trace data retention and export format unknown","Multi-agent visualization capabilities and supported topologies unknown","No built-in performance profiling (latency per step, token usage) documented","Breakpoint support and conditional debugging not documented"],"requires":["VS Code with Foundry Toolkit extension","Agent code written in supported framework (framework list unspecified)","F5 launch configuration in VS Code (auto-configured or manual setup unknown)"],"input_types":["agent code (Python/JavaScript)","agent inputs (prompts, tool definitions)"],"output_types":["execution trace (timeline of steps, tool calls, responses)","streaming responses (real-time model output)","structured logs (JSON trace data)"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-ms-windows-ai-studio-windows-ai-studio__cap_4","uri":"capability://data.processing.analysis.dataset.based.model.evaluation.with.built.in.and.custom.evaluators","name":"dataset-based model evaluation with built-in and custom evaluators","description":"Enables systematic model evaluation against datasets using a combination of built-in evaluators (F1 score, relevance, similarity, coherence) and custom evaluation criteria. Developers upload or reference datasets, define evaluation metrics, and run batch evaluations across models to compare performance. Results are displayed in a structured comparison view with metrics aggregation and per-sample analysis.","intents":["Benchmark model performance on domain-specific datasets","Compare multiple models objectively using standardized metrics","Identify model weaknesses on specific input types or edge cases","Validate model quality before production deployment"],"best_for":["ML engineers evaluating models for production use","Teams establishing model quality baselines","Researchers comparing model performance across variants"],"limitations":["Dataset format and size limits unknown (unclear supported formats: JSONL, CSV, Parquet, etc.)","Built-in evaluators limited to 4 types (F1, relevance, similarity, coherence) — no custom metric language or framework documented","Evaluation cost tracking for API-based models not documented","Batch evaluation performance and timeout limits unknown","No built-in statistical significance testing or confidence intervals"],"requires":["VS Code with Foundry Toolkit extension","Dataset file (format unspecified, likely JSONL or CSV)","Models to evaluate (must be accessible via API or local installation)","Evaluation criteria definition (UI or code-based, format unknown)"],"input_types":["dataset (JSONL, CSV, or other tabular format — unspecified)","model definitions (references to catalog models)","evaluation criteria (built-in metric names or custom definitions)"],"output_types":["evaluation results (metrics per model, per sample)","comparison reports (aggregated metrics, visualizations)","structured logs (evaluation traces)"],"categories":["data-processing-analysis","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-ms-windows-ai-studio-windows-ai-studio__cap_5","uri":"capability://automation.workflow.local.gpu.based.fine.tuning.with.cloud.fallback","name":"local gpu-based fine-tuning with cloud fallback","description":"Enables fine-tuning of models on local GPU hardware or via Azure Container Apps for cloud-based training. The extension abstracts away training infrastructure setup, handling data preparation, training loop orchestration, and model checkpointing. Developers specify a dataset, select a base model, configure training parameters (learning rate, epochs, batch size), and launch training either locally or in the cloud with progress monitoring within VS Code.","intents":["Fine-tune models on proprietary or domain-specific data without managing training infrastructure","Adapt pre-trained models to custom tasks with minimal code","Compare fine-tuned vs. base model performance on evaluation datasets","Deploy fine-tuned models back to local or cloud environments"],"best_for":["ML engineers adapting models to domain-specific tasks","Teams with limited ML infrastructure expertise","Organizations with GPU-constrained environments (cloud fallback)"],"limitations":["Local fine-tuning requires compatible GPU hardware (NVIDIA CUDA, AMD ROCm, or Intel Arc — support unspecified)","Supported model types for fine-tuning unknown (unclear if all catalog models support fine-tuning)","Training parameter exposure and customization scope unknown (unclear which hyperparameters are tunable)","Dataset format and size limits unknown","Cloud fine-tuning requires Azure subscription and incurs compute costs","No built-in distributed training or multi-GPU support documented","Fine-tuned model versioning and rollback not documented"],"requires":["VS Code with Foundry Toolkit extension","GPU hardware for local training (NVIDIA CUDA 11.8+, AMD ROCm, or Intel Arc — versions unspecified)","Training dataset (format unspecified, likely JSONL or CSV)","Base model accessible (from catalog or local installation)","Azure subscription for cloud fine-tuning (optional, for local fallback)"],"input_types":["dataset (JSONL, CSV, or other format — unspecified)","base model (reference from catalog)","training configuration (learning rate, epochs, batch size, optimizer — unspecified)"],"output_types":["fine-tuned model (ONNX, PyTorch, or other format — unspecified)","training logs (loss curves, metrics per epoch)","checkpoints (intermediate model states)"],"categories":["automation-workflow","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-ms-windows-ai-studio-windows-ai-studio__cap_6","uri":"capability://data.processing.analysis.model.quantization.and.format.conversion.with.onnx.support","name":"model quantization and format conversion with onnx support","description":"Provides automated model conversion and optimization workflows for transforming models between formats (Hugging Face to ONNX, quantization for edge deployment). The extension integrates with Hugging Face model hub, applies quantization techniques (int8, int4, or other precision reductions), and generates optimized models ready for local deployment via ONNX runtime or Ollama. Conversion progress and optimization metrics are displayed within VS Code.","intents":["Convert Hugging Face models to ONNX format for local deployment","Quantize models to reduce size and latency for edge/mobile deployment","Optimize models for specific hardware (CPU, GPU, NPU) using ONNX execution providers","Validate converted models against original for quality regression"],"best_for":["Developers deploying models to edge devices or resource-constrained environments","Teams optimizing model latency and memory footprint","Organizations standardizing on ONNX for cross-platform deployment"],"limitations":["Supported source formats limited to Hugging Face (no PyTorch, TensorFlow, JAX native support documented)","Quantization techniques and precision options unknown (unclear if int8, int4, float16 are all supported)","Conversion time and resource requirements unknown (unclear if conversion runs locally or in cloud)","Quality regression testing scope unknown (unclear which metrics are validated post-conversion)","ONNX opset version and execution provider support unknown","No built-in model compression techniques (pruning, distillation) documented"],"requires":["VS Code with Foundry Toolkit extension","Hugging Face model identifier (requires internet access to hub)","ONNX runtime (for local conversion, version unspecified)","Disk space for model files (size depends on model, unspecified)"],"input_types":["Hugging Face model identifier (string)","quantization configuration (precision level, technique — unspecified)","target hardware specification (CPU, GPU, NPU — unspecified)"],"output_types":["ONNX model file (.onnx)","quantized model (reduced precision)","conversion report (metrics, optimization results)"],"categories":["data-processing-analysis","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-ms-windows-ai-studio-windows-ai-studio__cap_7","uri":"capability://planning.reasoning.performance.tracing.and.metric.collection.for.agents","name":"performance tracing and metric collection for agents","description":"Collects and visualizes performance metrics during agent execution, including latency per step, token usage, API call costs, and resource consumption. Traces are captured automatically during F5 debugging or explicit trace collection, aggregated into a timeline view, and exported for analysis. Developers can identify bottlenecks, optimize expensive operations, and track cost implications of agent design choices.","intents":["Identify performance bottlenecks in agent workflows","Track API costs and token usage per agent execution","Optimize agent latency by analyzing per-step timing","Compare performance across model variants and configurations"],"best_for":["ML engineers optimizing agent performance and cost","Teams monitoring production agent efficiency","Developers making model selection decisions based on performance"],"limitations":["Trace data retention policy unknown (unclear if traces are persisted or ephemeral)","Metric granularity and completeness unknown (unclear which metrics are captured: token count, latency, cost, memory, etc.)","Cost tracking accuracy depends on model provider pricing data (may be outdated or incomplete)","No built-in alerting or threshold-based notifications documented","Export format and compatibility with external analysis tools unknown","Multi-agent trace correlation and aggregation scope unknown"],"requires":["VS Code with Foundry Toolkit extension","Agent code with tracing instrumentation (auto-instrumented or manual, unclear)","Model provider pricing data (for cost calculation, may require manual configuration)"],"input_types":["agent execution (captured during F5 debugging or explicit trace collection)","model provider configuration (for cost calculation)"],"output_types":["trace timeline (steps, latencies, tool calls)","aggregated metrics (total latency, token count, cost)","exported traces (format unspecified, likely JSON)"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-ms-windows-ai-studio-windows-ai-studio__cap_8","uri":"capability://data.processing.analysis.windows.ml.profiling.for.onnx.model.execution","name":"windows ml profiling for onnx model execution","description":"Provides CPU/GPU/NPU resource usage diagnostics and execution provider analysis for ONNX models running on Windows. The profiler captures Windows ML event traces, analyzes execution provider selection (CPU, GPU, TensorRT, CoreML), and reports resource consumption (memory, compute utilization). Results are displayed in VS Code with per-operation breakdown and optimization recommendations.","intents":["Profile ONNX model execution to identify resource bottlenecks","Validate execution provider selection (CPU vs GPU vs NPU) for optimal performance","Optimize model deployment on Windows devices with specific hardware","Debug model performance issues on target hardware"],"best_for":["Windows developers deploying ONNX models to edge devices","Teams optimizing model performance on specific Windows hardware","Organizations standardizing on Windows ML for inference"],"limitations":["Windows-only feature (no Linux, macOS support)","Requires Windows ML runtime (version unspecified, likely Windows 10 21H2+)","Profiling overhead and impact on model latency unknown","Supported execution providers unknown (likely CPU, GPU, TensorRT, but others unspecified)","Per-operation profiling granularity and accuracy unknown","No built-in optimization recommendations or automated tuning"],"requires":["Windows 10 21H2 or later (version unspecified)","VS Code with Foundry Toolkit extension","ONNX model file","Windows ML runtime (version unspecified)","Compatible hardware (GPU/NPU optional, CPU always available)"],"input_types":["ONNX model file","model inputs (test data for profiling)","execution provider configuration (CPU, GPU, NPU selection)"],"output_types":["profiling report (resource usage, per-operation breakdown)","execution provider analysis (selected provider, alternatives)","optimization recommendations (unspecified format)"],"categories":["data-processing-analysis","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-ms-windows-ai-studio-windows-ai-studio__cap_9","uri":"capability://tool.use.integration.mcp.tool.integration.for.agent.function.calling","name":"mcp tool integration for agent function calling","description":"Enables agents to invoke external tools via Model Context Protocol (MCP) integration, allowing structured function calling with schema-based tool definitions. Developers define tools as MCP resources, agents discover and invoke them with type-safe parameters, and results are returned to the agent for further processing. The extension manages tool registration, parameter validation, and error handling.","intents":["Extend agents with external tool capabilities (APIs, databases, file systems)","Enable agents to take actions in external systems (create tickets, send emails, update databases)","Provide type-safe function calling with schema validation","Integrate agents with existing enterprise systems and APIs"],"best_for":["Developers building agents that interact with external systems","Teams standardizing on MCP for tool integration","Organizations extending agents with custom business logic"],"limitations":["MCP tool discovery and registration mechanism unknown (unclear if automatic or manual)","Supported tool types and parameter schemas unknown (unclear if all JSON schema types are supported)","Tool error handling and retry logic not documented","Tool execution timeout and resource limits unknown","No built-in tool versioning or rollback capabilities documented","Tool authentication and credential management scope unknown"],"requires":["VS Code with Foundry Toolkit extension","MCP server implementation (language and framework unspecified)","Tool definitions (JSON schema format, unspecified)","Agent code with tool invocation support"],"input_types":["tool definitions (MCP resource definitions, JSON schema)","tool parameters (typed arguments matching schema)"],"output_types":["tool results (return values, structured data)","execution logs (tool invocation traces)"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":49,"verified":false,"data_access_risk":"high","permissions":["VS Code (version unspecified, likely 1.80+)","Windows OS (implied by 'Windows AI Studio' branding)","API keys for proprietary models (optional for local models via Ollama/ONNX)","VS Code with Foundry Toolkit extension installed","Selected model must be accessible (API key for cloud models, local installation for Ollama/ONNX)","Network connectivity for cloud-based models","VS Code with Foundry Toolkit extension","Multiple models configured and accessible","Agent code with orchestration logic (built-in patterns or custom, unclear)","Agent code in supported format"],"failure_modes":["Model catalog population requires valid API keys for proprietary providers (OpenAI, Anthropic, Google) or local installation (Ollama, ONNX)","No built-in model performance benchmarking — comparison is metadata-only (pricing, context window, capabilities)","Model availability depends on provider API status and network connectivity","Playground operates in-memory — no persistent conversation history or export functionality documented","Parameter tuning UI scope unknown (unclear which parameters are exposed for each model type)","Multi-modal input support limited to images and attachments (no video, audio, or custom formats documented)","No built-in cost tracking for API calls made during experimentation","Supported orchestration patterns unknown (unclear if all patterns mentioned are implemented)","Model selection criteria and comparison metrics scope unknown","Latency impact of parallel execution and comparison not documented","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.8,"quality":0.34,"ecosystem":0.35000000000000003,"match_graph":0.25,"freshness":0.75,"weights":{"adoption":0.25,"quality":0.25,"ecosystem":0.15,"match_graph":0.23,"freshness":0.12}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:34.803Z","last_scraped_at":"2026-05-03T15:20:29.937Z","last_commit":null},"community":{"stars":null,"forks":null,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=foundry-toolkit-for-vs-code","compare_url":"https://unfragile.ai/compare?artifact=foundry-toolkit-for-vs-code"}},"signature":"mo4MgeQ/q7CU8NkgEAq3SETl32O16IpMfdtwaXSSDZEJOQHSU84PLjOx+IqQKBcZisPZMLB/FrwQFq+rdKlSDQ==","signedAt":"2026-06-21T18:38:38.290Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/foundry-toolkit-for-vs-code","artifact":"https://unfragile.ai/foundry-toolkit-for-vs-code","verify":"https://unfragile.ai/api/v1/verify?slug=foundry-toolkit-for-vs-code","publicKey":"https://unfragile.ai/api/v1/trust-passport-public-key","spec":"https://unfragile.ai/trust","schema":"https://unfragile.ai/schema.json","docs":"https://unfragile.ai/docs"}}