{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"vscode-harshagarwal1012-openclaude-vscode","slug":"openclaude-vs-code","name":"OpenClaude VS Code","type":"extension","url":"https://marketplace.visualstudio.com/items?itemName=HarshAgarwal1012.openclaude-vscode","page_url":"https://unfragile.ai/openclaude-vs-code","categories":["code-editors"],"tags":["agent","ai","claude","json","keybindings","mcp","sonnet"],"pricing":{"model":"freemium","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"vscode-harshagarwal1012-openclaude-vscode__cap_0","uri":"capability://text.generation.language.multi.provider.llm.chat.with.runtime.provider.switching","name":"multi-provider llm chat with runtime provider switching","description":"Provides a VS Code sidebar chat panel that streams responses from 8+ LLM providers (OpenAI, Anthropic, Google Gemini, Ollama, AWS Bedrock, GitHub Models, and OpenAI-compatible custom endpoints) with runtime provider switching via `/provider` slash command or UI badge. The extension wraps the OpenClaude CLI, delegating model inference to the CLI process while rendering markdown-formatted streaming responses with syntax-highlighted code blocks in the native VS Code chat interface. Provider credentials are configured via environment variables (OPENAI_API_KEY, GOOGLE_API_KEY, etc.) or interactive setup commands.","intents":["Switch between Claude, GPT-4, Gemini, and local Ollama models without leaving VS Code","Use a local Ollama model for offline coding without API keys","Compare responses from different LLM providers on the same coding problem","Configure a custom OpenAI-compatible endpoint (e.g., vLLM, LocalAI) for inference"],"best_for":["developers evaluating multiple LLM providers for coding tasks","teams with heterogeneous model preferences (some prefer Claude, others GPT-4)","organizations running local Ollama instances and wanting VS Code integration","builders prototyping multi-model comparison workflows"],"limitations":["Requires OpenClaude CLI as external dependency — extension cannot function without `npm install -g @gitlawb/openclaude`","All non-local providers require internet connectivity and valid API credentials","Provider switching is manual via command/UI — no automatic fallback if primary provider fails","No built-in provider load balancing or cost optimization across multiple accounts","Token counting in status bar is approximate and may not match actual provider billing"],"requires":["Node.js 14+ (for OpenClaude CLI installation)","OpenClaude CLI installed globally via npm","VS Code 1.60+ (inferred from extension API usage)","API key for at least one provider (OpenAI, Anthropic, Google, etc.) OR local Ollama running on port 11434","Environment variables configured for chosen provider(s)"],"input_types":["natural language text (chat messages)","code snippets (via @-mention context)","file paths (via @-mention syntax)"],"output_types":["markdown-formatted text with syntax highlighting","code blocks with language-specific highlighting","streaming text responses","tool invocation metadata (collapsible blocks)"],"categories":["text-generation-language","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-harshagarwal1012-openclaude-vscode__cap_1","uri":"capability://memory.knowledge.selective.file.folder.line.range.context.inclusion.via.mention.syntax","name":"selective file/folder/line-range context inclusion via @-mention syntax","description":"Implements a @-mention system (similar to Slack or GitHub) allowing developers to explicitly include file contents, entire folders, or specific line ranges in chat context without automatic project-wide scanning. When a user types `@filename.js`, `@folder/`, or `@file.js:10-20`, the extension resolves the path relative to the workspace root, reads the file contents, and injects them into the LLM context window. This approach avoids token waste on irrelevant files and gives developers fine-grained control over context scope, critical for large codebases where full project indexing would exceed token limits.","intents":["Ask the AI about a specific file without including the entire project","Get code review feedback on lines 10-20 of a file without sending the whole file","Include all files in a folder (e.g., `@src/components/`) to discuss architectural patterns","Reduce token usage by explicitly selecting only relevant files for a query"],"best_for":["developers working in large codebases (>10k LOC) where full project context exceeds token limits","teams with strict API cost constraints wanting fine-grained context control","developers debugging specific modules and wanting to isolate context to relevant files","builders integrating LLMs into existing VS Code workflows without disrupting file access patterns"],"limitations":["No automatic dependency resolution — if you mention `@file.js` but it imports from `utils.js`, the AI won't see `utils.js` unless explicitly mentioned","Line-range syntax requires manual calculation — no IDE-assisted range selection (e.g., select lines in editor, auto-populate @mention)","Folder inclusion (@folder/) likely includes all files recursively, no granular filtering by file type or size","No caching of frequently-mentioned files — each mention re-reads from disk","Path resolution behavior unclear for symlinks, relative imports, or files outside workspace root"],"requires":["VS Code workspace with files accessible to the extension process","Knowledge of relative file paths from workspace root","Correct @-mention syntax (exact format not fully documented)"],"input_types":["file paths (relative to workspace root)","folder paths","line ranges (e.g., filename:10-20)"],"output_types":["file contents injected into LLM context","line-range excerpts","folder file listings (inferred)"],"categories":["memory-knowledge","code-generation-editing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-harshagarwal1012-openclaude-vscode__cap_10","uri":"capability://tool.use.integration.mcp.model.context.protocol.server.integration.and.plugin.management","name":"mcp (model context protocol) server integration and plugin management","description":"The extension integrates with the Model Context Protocol (MCP), an open standard for extending LLM context with external data sources and tools. The extension includes an MCP plugin manager that allows developers to install and configure MCP servers (e.g., for accessing databases, APIs, file systems, or custom knowledge bases). When an MCP server is enabled, the extension automatically includes its resources and tools in the LLM's context, allowing the AI to query external data sources or invoke external tools. This architecture decouples context sources from the extension itself, enabling extensibility without modifying the extension code.","intents":["Connect the AI to external data sources (databases, APIs, knowledge bases) without custom integration","Extend the AI's capabilities with custom tools via MCP servers","Use open-source MCP servers (e.g., for GitHub, Slack, Notion) without building custom integrations","Build custom MCP servers for organization-specific tools and data sources"],"best_for":["developers wanting to extend AI capabilities beyond code generation","teams with custom tools and data sources wanting to integrate with AI","builders creating MCP servers for organization-specific use cases","organizations adopting MCP as a standard for LLM integrations"],"limitations":["MCP is a relatively new standard — ecosystem of available servers is limited","MCP server installation and configuration is not documented — unclear if it's via npm, git, or a package manager","No built-in MCP server discovery or marketplace — developers must know which servers to install","MCP servers add latency to LLM context retrieval — unclear if context is cached or fetched on every query","Security implications of MCP servers are unclear — no documented sandboxing or permission model","Debugging MCP server failures is difficult — error messages may be opaque"],"requires":["MCP server(s) installed and running (local or remote)","MCP plugin manager in the extension (built-in)","Configuration of MCP server endpoints and credentials"],"input_types":["MCP server URLs or local paths","MCP server credentials (API keys, etc.)","MCP resource queries (e.g., database queries, API calls)"],"output_types":["MCP resource data (injected into LLM context)","MCP tool results (from tool invocations)","plugin manager UI"],"categories":["tool-use-integration","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-harshagarwal1012-openclaude-vscode__cap_11","uri":"capability://automation.workflow.onboarding.walkthrough.for.new.user.setup","name":"onboarding walkthrough for new user setup","description":"The extension includes an interactive onboarding walkthrough that guides new users through initial setup, including provider selection, API key configuration, keybinding explanation, and feature overview. The walkthrough is likely triggered on first installation and can be re-triggered via a command. It provides a structured, step-by-step introduction to the extension's capabilities, reducing the learning curve and setup friction. The walkthrough may include interactive examples (e.g., 'try asking the AI a question') to familiarize users with the chat interface.","intents":["Get started with OpenClaude without reading documentation","Understand available features and keybindings through guided examples","Configure the first LLM provider without manual setup","Learn best practices for using the extension (e.g., @-mention syntax, permission modes)"],"best_for":["new users unfamiliar with AI coding assistants","non-technical developers wanting guided setup","teams onboarding developers to the extension","organizations wanting to standardize extension usage"],"limitations":["Onboarding walkthrough content and UX are not documented","Unclear whether walkthrough is skippable or mandatory","No personalization based on user experience level (e.g., advanced users may find it tedious)","Walkthrough may become outdated if extension features change","No analytics on walkthrough completion or dropout rates"],"requires":["VS Code 1.60+ (for webview API used in walkthrough)","First-time extension installation (walkthrough likely triggered automatically)"],"input_types":["user interactions (button clicks, text input) during walkthrough"],"output_types":["walkthrough UI (webview or modal)","configuration saved from walkthrough","example chat messages"],"categories":["automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-harshagarwal1012-openclaude-vscode__cap_12","uri":"capability://automation.workflow.keyboard.shortcut.integration.for.quick.access.and.context.insertion","name":"keyboard shortcut integration for quick access and context insertion","description":"The extension provides global keyboard shortcuts for common actions: `Cmd+Escape` (Mac) / `Ctrl+Escape` (Windows/Linux) to open/focus the chat panel, and `Cmd+Shift+Escape` / `Ctrl+Shift+Escape` to open the chat in a new tab. Additionally, `Alt+[key]` shortcuts enable quick @-mention insertion (exact keys not fully documented). These shortcuts are registered with VS Code's keybinding system and can be customized by users via the keybindings.json file. The shortcuts provide quick access without requiring mouse navigation or command palette usage.","intents":["Quickly open the chat panel without using the mouse or command palette","Switch focus between editor and chat panel using keyboard","Open chat in a new tab for side-by-side editing and AI assistance","Insert @-mentions quickly without typing the full syntax"],"best_for":["keyboard-driven developers wanting to minimize mouse usage","developers with high context-switching frequency between editor and chat","power users optimizing for speed and efficiency","developers with accessibility needs (keyboard-only navigation)"],"limitations":["Default shortcuts may conflict with other VS Code extensions or user keybindings","Alt+[key] shortcuts for @-mention insertion are not fully documented — unclear which keys are supported","No customization UI for keybindings — users must edit keybindings.json manually","Shortcuts are global, not context-aware (e.g., can't have different shortcuts in different file types)","No macro support for complex multi-step shortcuts"],"requires":["VS Code 1.60+ (for keybinding API)","Keyboard (obviously)","Knowledge of default shortcuts or willingness to customize"],"input_types":["keyboard input (key combinations)"],"output_types":["chat panel focus/open","new chat tab","@-mention insertion"],"categories":["automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-harshagarwal1012-openclaude-vscode__cap_2","uri":"capability://code.generation.editing.ai.proposed.code.changes.with.native.diff.viewer.and.accept.reject.workflow","name":"ai-proposed code changes with native diff viewer and accept/reject workflow","description":"When the LLM generates code changes, the extension renders them in VS Code's native diff viewer (side-by-side or unified diff format), allowing developers to review proposed edits before applying them. The workflow is: AI generates code → extension parses response for code blocks → creates a temporary file or diff representation → opens native VS Code diff UI → developer clicks 'Accept' (applies changes) or 'Reject' (discards). This integrates seamlessly with VS Code's built-in diff viewer, avoiding custom UI and leveraging familiar editor affordances.","intents":["Review AI-generated code changes before applying them to avoid breaking changes","See exactly what lines the AI modified (additions, deletions, modifications highlighted)","Accept partial changes and reject others in the same response","Maintain code review discipline by forcing explicit approval of AI edits"],"best_for":["developers prioritizing code safety and wanting explicit approval gates on AI edits","teams with code review policies that require human sign-off on all changes","solo developers wanting to catch AI hallucinations before they hit the codebase","codebases with strict quality standards where blind AI application is unacceptable"],"limitations":["Diff viewer only shows changes proposed in a single LLM response — multi-turn refinements require multiple diff reviews","No automatic conflict detection if developer manually edited the file between AI suggestion and acceptance","Unclear whether extension supports partial acceptance (accepting some hunks, rejecting others) or all-or-nothing","No integration with git staging — accepted changes are applied directly to working directory, not staged","Diff viewer may struggle with large changes (>1000 lines) due to VS Code rendering performance"],"requires":["VS Code 1.60+ (for diff viewer API)","File must be writable in the workspace","LLM response must contain code blocks in markdown format (triple-backtick syntax)"],"input_types":["markdown code blocks from LLM response","file paths for target files"],"output_types":["VS Code diff viewer UI","applied file changes (on accept)","discarded changes (on reject)"],"categories":["code-generation-editing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-harshagarwal1012-openclaude-vscode__cap_3","uri":"capability://memory.knowledge.multi.turn.conversation.history.with.fork.resume.and.checkpoint.capabilities","name":"multi-turn conversation history with fork, resume, and checkpoint capabilities","description":"The extension maintains a persistent conversation history for each chat session, allowing developers to browse past conversations, resume interrupted sessions, and fork conversations at any point to explore alternative paths. The architecture stores conversation metadata (messages, model used, provider, timestamp) locally or in extension storage, enabling quick retrieval without re-querying the LLM. Forking creates a branch point in the conversation tree, allowing developers to ask 'what if' questions without losing the original conversation thread. This is similar to ChatGPT's conversation management but integrated into VS Code's sidebar.","intents":["Resume a coding session from yesterday without losing context or conversation history","Fork a conversation to explore an alternative solution path while keeping the original intact","Review past AI suggestions and code changes without re-running the same queries","Compare multiple solution approaches by forking and exploring different branches"],"best_for":["developers working on long-running projects requiring multi-day context continuity","teams exploring multiple solution approaches and wanting to preserve decision trees","developers wanting to audit AI suggestions over time (compliance, learning)","builders prototyping iterative AI-assisted workflows"],"limitations":["Conversation history is stored locally in VS Code extension storage — not synced across devices or machines","No built-in export of conversation history (unclear if JSON/markdown export is supported)","Forking creates separate conversation branches, but no UI visualization of the conversation tree (unclear if tree view exists)","Resuming a conversation may not preserve the exact LLM state if provider or model changed between sessions","Storage limits unknown — very long conversations may hit VS Code extension storage quotas","No collaborative conversation sharing — history is per-user, not team-accessible"],"requires":["VS Code extension storage API (built-in, no external dependency)","Conversation metadata serialization format (implementation details unknown)"],"input_types":["conversation IDs or timestamps","fork point identifiers"],"output_types":["conversation list UI","resumed chat context","forked conversation branches"],"categories":["memory-knowledge","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-harshagarwal1012-openclaude-vscode__cap_4","uri":"capability://text.generation.language.streaming.response.rendering.with.markdown.and.syntax.highlighted.code.blocks","name":"streaming response rendering with markdown and syntax-highlighted code blocks","description":"As the LLM generates tokens, the extension streams them to the VS Code chat panel in real-time, parsing markdown syntax and rendering code blocks with language-specific syntax highlighting. The implementation uses a markdown parser (likely a lightweight library) to identify code fences (triple backticks with language specifiers), extract the language identifier, and apply VS Code's built-in syntax highlighter for that language. Streaming is non-blocking — the UI updates incrementally as tokens arrive, providing immediate feedback to the developer. The extension also supports interrupting the stream via a 'Stop' button.","intents":["See AI responses appear in real-time without waiting for full generation","Read code suggestions with proper syntax highlighting for readability","Stop a response mid-generation if the AI is going off-track","Copy code blocks directly from the chat panel with correct formatting"],"best_for":["developers valuing low-latency feedback and real-time interaction","teams with slow API connections wanting to see partial responses quickly","developers reading long responses and wanting to start reading before generation completes","builders integrating streaming LLM responses into VS Code without custom rendering"],"limitations":["Streaming rendering may cause UI jank if markdown parsing is done on the main thread (performance impact unknown)","Syntax highlighting is limited to languages VS Code natively supports — custom or obscure languages may not highlight","Interrupting a stream stops the response but doesn't roll back partial changes to the file (if auto-apply is enabled)","Markdown parsing may fail on malformed code blocks, potentially breaking rendering","No support for streaming structured data (JSON, tables) — only markdown and code blocks"],"requires":["VS Code 1.60+ (for webview streaming API)","LLM provider supporting streaming (OpenAI, Anthropic, Ollama all support this)","Network connection with reasonable latency (streaming is ineffective on high-latency connections)"],"input_types":["streaming token stream from LLM provider","markdown-formatted text"],"output_types":["rendered markdown in VS Code chat panel","syntax-highlighted code blocks","copyable code snippets"],"categories":["text-generation-language","code-generation-editing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-harshagarwal1012-openclaude-vscode__cap_5","uri":"capability://tool.use.integration.tool.invocation.visualization.and.execution.tracking","name":"tool invocation visualization and execution tracking","description":"When the LLM calls external tools (e.g., file system operations, API calls, code execution), the extension renders these invocations as collapsible blocks in the chat panel, showing the tool name, input parameters, and execution result. The implementation parses tool-calling responses from the LLM (likely using OpenAI's function-calling format or similar), executes the tool via the OpenClaude CLI or native handlers, and streams the result back into the chat context. This provides transparency into what the AI is doing beyond just generating text — developers can see file reads, API calls, and other side effects.","intents":["Understand what files the AI read or modified during code generation","Debug AI decisions by seeing what tools it invoked and what results it received","Approve or reject tool invocations before they execute (if approval workflow exists)","Audit AI actions for security or compliance purposes"],"best_for":["developers wanting transparency into AI reasoning and side effects","teams with security policies requiring audit trails of AI actions","builders debugging AI agent behavior and tool-calling logic","developers concerned about unintended file modifications or API calls"],"limitations":["Tool invocation approval workflow is unclear — unclear if developers can reject tool calls before execution","Supported tools are unknown — documentation doesn't enumerate which tools the AI can invoke","Tool execution happens via OpenClaude CLI, adding latency and potential failure modes","No built-in rate limiting or quota enforcement for tool invocations","Tool results are injected back into LLM context, potentially creating feedback loops if tools fail"],"requires":["OpenClaude CLI with tool support (implementation details unknown)","LLM provider supporting function calling (OpenAI, Anthropic, etc.)","Appropriate permissions for tools (e.g., file system access for file operations)"],"input_types":["tool-calling responses from LLM","tool parameters (JSON or structured format)"],"output_types":["collapsible tool invocation blocks in chat UI","tool execution results","error messages if tools fail"],"categories":["tool-use-integration","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-harshagarwal1012-openclaude-vscode__cap_6","uri":"capability://data.processing.analysis.real.time.token.count.and.cost.estimation.in.status.bar","name":"real-time token count and cost estimation in status bar","description":"The extension displays a live token count and estimated cost in the VS Code status bar, updating as the user types messages and the LLM generates responses. The implementation likely tokenizes the current chat context using the selected LLM's tokenizer (e.g., OpenAI's cl100k_base for GPT-4, Anthropic's tokenizer for Claude) and multiplies by the provider's per-token pricing to estimate cost. This helps developers stay aware of API spending and avoid expensive queries. The status bar widget is clickable, potentially opening a cost breakdown or provider settings.","intents":["Monitor API costs in real-time to avoid surprise bills","Decide whether to include large files in context based on token cost","Compare cost efficiency of different models (GPT-4 vs GPT-4o-mini)","Set spending alerts or limits based on token usage"],"best_for":["developers paying out-of-pocket for API usage and wanting cost visibility","teams with API budgets and needing to track spending","builders optimizing prompt efficiency and wanting to measure token savings","developers experimenting with multiple models and comparing costs"],"limitations":["Token counting is approximate — actual provider billing may differ due to rounding, special tokens, or pricing changes","Cost estimation assumes static pricing — doesn't account for volume discounts or dynamic pricing","No built-in spending alerts or hard limits — developers must manually monitor","Tokenizer may be outdated if LLM provider changes tokenization scheme","Cost display is per-message, not cumulative per session or per day","Unclear whether cost includes both input and output tokens or just input"],"requires":["Tokenizer library for selected LLM provider (bundled with extension or OpenClaude CLI)","Current pricing data for each provider (must be updated when providers change pricing)"],"input_types":["chat messages (text)","file context (@-mentions)","LLM responses"],"output_types":["token count (integer)","estimated cost (USD or other currency)","status bar widget"],"categories":["data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-harshagarwal1012-openclaude-vscode__cap_7","uri":"capability://automation.workflow.git.worktree.support.for.parallel.ai.assisted.sessions","name":"git worktree support for parallel ai-assisted sessions","description":"The extension integrates with Git worktrees, allowing developers to run separate OpenClaude sessions for different branches or worktrees without context collision. When a developer switches to a different Git worktree (e.g., `git worktree add ../feature-branch`), the extension detects the change and isolates the chat session, conversation history, and context to that worktree. This enables parallel development workflows where multiple team members or the same developer can work on different branches with independent AI assistance without mixing context.","intents":["Work on multiple feature branches simultaneously with separate AI sessions per branch","Avoid context pollution when switching between branches (e.g., AI suggesting code from the wrong branch)","Maintain separate conversation histories for different features or bug fixes","Enable team members to work on different worktrees without interfering with each other's AI sessions"],"best_for":["developers using Git worktrees for parallel development workflows","teams with multiple developers working on different branches simultaneously","large projects where context isolation between branches is critical","builders integrating AI assistance into multi-branch development workflows"],"limitations":["Requires Git worktrees to be set up — not all teams use worktrees (many use branch switching instead)","Worktree detection mechanism is unclear — may not handle all edge cases (e.g., nested worktrees, symlinks)","No automatic context sharing between worktrees — if you want to reference code from another branch, you must manually @-mention it","Conversation history is per-worktree, not per-branch — switching back to a worktree may not restore the exact previous session state","Performance impact of worktree detection is unknown"],"requires":["Git 2.7+ (for worktree support)","Git worktrees already created in the workspace","VS Code workspace configured to detect worktree changes"],"input_types":["git worktree paths","branch names"],"output_types":["isolated chat sessions per worktree","separate conversation histories","worktree-specific context"],"categories":["automation-workflow","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-harshagarwal1012-openclaude-vscode__cap_8","uri":"capability://safety.moderation.permission.modes.for.controlling.ai.edit.authorization.levels","name":"permission modes for controlling ai edit authorization levels","description":"The extension provides 5 permission modes (Default, Plan, Accept Edits, Bypass, Don't Ask) that control how the AI can modify files and what approval is required. While the exact behavior of each mode is not fully documented, the architecture likely works as follows: each mode defines a threshold for automatic approval (e.g., 'Bypass' auto-applies all edits, 'Accept Edits' requires explicit approval, 'Don't Ask' disables edits entirely). The extension checks the current permission mode before applying AI-suggested changes and enforces the corresponding workflow. This provides fine-grained control over AI autonomy, from fully autonomous to fully manual.","intents":["Allow the AI to auto-apply small, low-risk edits while requiring approval for large changes","Disable AI file modifications entirely for safety-critical code","Require explicit approval for all AI edits in a specific file or project","Gradually increase AI autonomy as trust in the system grows"],"best_for":["teams with varying trust levels in AI-generated code","projects with safety-critical code requiring human review of all changes","developers wanting to experiment with different autonomy levels","organizations with compliance requirements for code change approval"],"limitations":["Exact behavior of each permission mode is not documented — unclear what 'Plan' vs 'Accept Edits' means","No granular control per file or folder — permission modes are global or per-session","No integration with code review workflows (e.g., requiring PR approval before AI changes are committed)","Unclear whether permission modes apply to tool invocations or only file edits","No audit trail of which permission mode was active when an edit was applied"],"requires":["VS Code extension configuration or settings UI to select permission mode","Clear documentation of what each mode does (currently missing)"],"input_types":["permission mode selection (enum: Default, Plan, Accept Edits, Bypass, Don't Ask)","AI-suggested changes"],"output_types":["applied file changes (if approved)","approval dialogs (if required)","rejected changes (if not approved)"],"categories":["safety-moderation","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"vscode-harshagarwal1012-openclaude-vscode__cap_9","uri":"capability://tool.use.integration.interactive.provider.configuration.via.slash.command","name":"interactive provider configuration via slash command","description":"The extension provides a `/provider` slash command that opens an interactive setup wizard for configuring LLM providers without manually editing environment variables. The wizard likely walks through provider selection (OpenAI, Anthropic, Gemini, etc.), API key entry, model selection, and optional advanced settings (base URL for custom endpoints, temperature, max tokens). The configuration is stored in VS Code's extension settings or a local config file, and the extension reads these settings at startup. This lowers the barrier to entry for non-technical users and reduces configuration errors from typos in environment variables.","intents":["Set up an LLM provider without touching environment variables or terminal","Switch providers interactively without restarting VS Code","Configure a custom OpenAI-compatible endpoint (e.g., vLLM) via the UI","Store API keys securely in VS Code's credential storage (if supported)"],"best_for":["non-technical developers unfamiliar with environment variables","teams wanting to standardize provider configuration across developers","developers frequently switching between providers","organizations with custom LLM endpoints wanting easy configuration"],"limitations":["Interactive setup wizard UI/UX is not documented — unclear if it's a form, dropdown menu, or text input","Unclear whether API keys are stored securely in VS Code's credential storage or in plaintext config files","No validation of API keys during setup — invalid keys may only be detected when making the first API call","Slash command may conflict with other extensions using `/` for commands","No support for environment variable interpolation (e.g., `/provider` can't read from `$OPENAI_API_KEY`)"],"requires":["VS Code 1.60+ (for command palette and settings API)","API key for the chosen provider (obtained from provider's dashboard)"],"input_types":["provider selection (dropdown or text input)","API key (text input, ideally masked)","model name (dropdown or text input)","optional: base URL for custom endpoints"],"output_types":["VS Code extension settings (stored in settings.json or credential storage)","confirmation message in chat panel"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":38,"verified":false,"data_access_risk":"high","permissions":["Node.js 14+ (for OpenClaude CLI installation)","OpenClaude CLI installed globally via npm","VS Code 1.60+ (inferred from extension API usage)","API key for at least one provider (OpenAI, Anthropic, Google, etc.) OR local Ollama running on port 11434","Environment variables configured for chosen provider(s)","VS Code workspace with files accessible to the extension process","Knowledge of relative file paths from workspace root","Correct @-mention syntax (exact format not fully documented)","MCP server(s) installed and running (local or remote)","MCP plugin manager in the extension (built-in)"],"failure_modes":["Requires OpenClaude CLI as external dependency — extension cannot function without `npm install -g @gitlawb/openclaude`","All non-local providers require internet connectivity and valid API credentials","Provider switching is manual via command/UI — no automatic fallback if primary provider fails","No built-in provider load balancing or cost optimization across multiple accounts","Token counting in status bar is approximate and may not match actual provider billing","No automatic dependency resolution — if you mention `@file.js` but it imports from `utils.js`, the AI won't see `utils.js` unless explicitly mentioned","Line-range syntax requires manual calculation — no IDE-assisted range selection (e.g., select lines in editor, auto-populate @mention)","Folder inclusion (@folder/) likely includes all files recursively, no granular filtering by file type or size","No caching of frequently-mentioned files — each mention re-reads from disk","Path resolution behavior unclear for symlinks, relative imports, or files outside workspace root","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.38,"quality":0.35,"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":"2026-05-03T15:20:33.198Z","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=openclaude-vs-code","compare_url":"https://unfragile.ai/compare?artifact=openclaude-vs-code"}},"signature":"VNSQmZfCcaEcCjghQSJvAzsA6Y/5QhS9bG5ivE7gjmLuJXGl4mzhX6yviAF9BL+ggMqsH695eK9NE3B4HBfOAA==","signedAt":"2026-06-21T10:09:02.991Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/openclaude-vs-code","artifact":"https://unfragile.ai/openclaude-vs-code","verify":"https://unfragile.ai/api/v1/verify?slug=openclaude-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"}}