{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"github-cowork-os--cowork-os","slug":"cowork-os--cowork-os","name":"CoWork-OS","type":"agent","url":"https://github.com/CoWork-OS/CoWork-OS","page_url":"https://unfragile.ai/cowork-os--cowork-os","categories":["ai-agents"],"tags":["ai-assistants","ai-automation","claude","desktop-app","discord-bot","electron","gemini","guardrails","llm","local-first","macos","mcp","model-context-protocol","openai","personal-ai","security","self-hosted","slack-bot","telegram","whatsapp-bot"],"pricing":{"model":"open_source","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"github-cowork-os--cowork-os__cap_0","uri":"capability://tool.use.integration.multi.channel.agent.deployment.with.unified.message.routing","name":"multi-channel agent deployment with unified message routing","description":"Deploys a single AI agent across WhatsApp, Telegram, Discord, Slack, and iMessage through a unified message routing layer that normalizes incoming messages into a common schema, routes them through the agent pipeline, and formats responses back to each platform's native API format. Uses adapter pattern with platform-specific SDK integrations (Twilio for WhatsApp, Telegram Bot API, Discord.js, Slack Bolt, iMessage via native macOS APIs) that translate between platform message formats and internal message objects.","intents":["Deploy a single AI agent to multiple chat platforms without rewriting logic for each channel","Manage agent conversations across WhatsApp, Telegram, Discord, Slack, and iMessage from one codebase","Route user messages from any platform through consistent agent processing pipeline","Maintain conversation context and user identity across different messaging platforms"],"best_for":["Teams building multi-platform AI assistants without platform-specific rewrites","Developers wanting to reach users where they already communicate (WhatsApp, Slack, Discord)","Organizations needing unified agent deployment across enterprise and consumer channels"],"limitations":["Platform-specific features (rich media, interactive buttons) require custom adapter code per channel","Rate limiting and quota management must be handled per-platform; no unified rate limiter","Message ordering guarantees depend on underlying platform reliability; no built-in deduplication","iMessage support limited to macOS; requires running on Apple hardware for that channel"],"requires":["TypeScript runtime (Node.js 18+)","Platform API credentials (Twilio account for WhatsApp, Telegram Bot Token, Discord Bot Token, Slack Bot Token)","macOS system for iMessage channel support","Network connectivity to platform APIs or self-hosted message broker"],"input_types":["text messages","media attachments (platform-dependent)","user metadata (ID, name, profile)"],"output_types":["text responses","formatted messages per platform","structured metadata (user ID, channel, timestamp)"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cowork-os--cowork-os__cap_1","uri":"capability://tool.use.integration.multi.provider.llm.abstraction.with.provider.agnostic.prompting","name":"multi-provider llm abstraction with provider-agnostic prompting","description":"Abstracts Claude, GPT, Gemini, and Ollama behind a unified provider interface that accepts model-agnostic prompts and routes them to the appropriate provider's API with format translation. Handles provider-specific differences in API contracts (message format, parameter names, response structure) through a provider registry pattern, allowing agents to switch models or providers without changing prompt logic. Supports streaming and non-streaming responses with unified callback handling.","intents":["Switch between Claude, GPT, Gemini, and Ollama without rewriting agent prompts","Compare model outputs across providers for the same prompt","Fall back to alternative providers if primary provider is unavailable or rate-limited","Use local Ollama models for privacy-sensitive tasks while using cloud providers for others"],"best_for":["Developers building provider-agnostic AI agents","Teams evaluating multiple LLM providers without code refactoring","Organizations with privacy requirements wanting to mix local (Ollama) and cloud models","Builders needing fallback strategies across multiple LLM providers"],"limitations":["Model-specific capabilities (vision, function calling) require provider-specific code paths despite abstraction","Prompt engineering optimizations for one model may not transfer to another; abstraction hides model-specific tuning","Cost tracking and quota management per provider must be implemented separately","Streaming response handling adds ~50-100ms latency due to abstraction layer translation"],"requires":["API keys for desired providers (OpenAI, Anthropic, Google, or local Ollama instance)","TypeScript runtime (Node.js 18+)","Network connectivity to cloud providers or local Ollama server"],"input_types":["text prompts","message arrays with role/content structure","system instructions"],"output_types":["text completions","streaming token streams","structured metadata (model, tokens used, provider)"],"categories":["tool-use-integration","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cowork-os--cowork-os__cap_10","uri":"capability://tool.use.integration.desktop.integration.with.native.system.apis.file.access.clipboard.notifications","name":"desktop integration with native system apis (file access, clipboard, notifications)","description":"Provides agents with access to native macOS system capabilities through Electron bridge: file system access (read/write files), clipboard operations (read/write), system notifications, and native dialogs. Implements sandboxed access where agents declare required system permissions upfront, and runtime validates each system call against declared permissions. Uses Electron IPC (Inter-Process Communication) to safely bridge agent process and native APIs.","intents":["Allow agents to read and write files on user's computer for document processing","Enable agents to access clipboard for quick data exchange","Send system notifications to alert user of agent actions or results","Integrate agents with native macOS workflows (file monitoring, system events)"],"best_for":["Desktop AI assistants on macOS requiring local file access","Personal productivity agents that work with user's documents and files","Developers building native macOS applications with AI capabilities","Teams wanting to avoid cloud storage for sensitive documents"],"limitations":["Desktop integration limited to macOS; no native Windows or Linux support (Docker alternative available)","File access permissions are coarse-grained; no per-file or per-directory granularity","Clipboard operations are synchronous; large clipboard operations may block UI","System notifications are basic; no rich interactive notifications","Electron adds significant memory overhead (~200MB per instance) vs. headless deployment"],"requires":["macOS 10.15+","Electron runtime (included in CoWork-OS desktop app)","User permission grants for file system and clipboard access"],"input_types":["file paths","clipboard read/write requests","notification content"],"output_types":["file contents","clipboard data","notification delivery confirmation"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cowork-os--cowork-os__cap_11","uri":"capability://safety.moderation.audit.logging.and.compliance.reporting.with.structured.event.capture","name":"audit logging and compliance reporting with structured event capture","description":"Captures all agent actions, tool calls, capability requests, and security decisions as structured audit logs with timestamps, user IDs, agent IDs, and outcomes. Stores logs in queryable format (JSON, database) with configurable retention policies. Generates compliance reports (who did what, when, why) for security investigations and regulatory audits. Supports log export in standard formats (CSV, JSON) for external analysis.","intents":["Investigate security incidents by reviewing agent action history","Generate audit trails for compliance requirements (SOC 2, HIPAA, GDPR)","Track which agents accessed which resources for access control audits","Monitor for suspicious patterns (unusual access, repeated failures)"],"best_for":["Organizations with compliance requirements (SOC 2, HIPAA, GDPR, etc.)","Security teams investigating incidents or suspicious activity","Teams needing to prove audit trails for regulatory audits","Developers building multi-tenant platforms with per-tenant audit logs"],"limitations":["Audit logging adds ~10-20ms latency per action due to log writing","Log storage grows rapidly; 1000 requests/day = ~30GB/year with full event capture","No built-in log analysis; requires external tools (ELK, Splunk) for pattern detection","Structured logging requires careful schema design; schema changes require migration","Log retention policies must balance compliance requirements with storage costs"],"requires":["TypeScript runtime (Node.js 18+)","Persistent storage for logs (file system, database, or log aggregation service)","Log retention policy configuration"],"input_types":["agent actions (tool calls, capability requests)","security decisions (allow/deny)","user and agent metadata"],"output_types":["structured audit log entries (JSON)","compliance reports (CSV, JSON)","log queries and analysis results"],"categories":["safety-moderation","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cowork-os--cowork-os__cap_2","uri":"capability://safety.moderation.security.first.agent.sandboxing.with.capability.based.access.control","name":"security-first agent sandboxing with capability-based access control","description":"Enforces security through capability-based access control where agents declare required permissions (file access, network calls, tool execution) upfront, and the runtime validates each agent action against declared capabilities before execution. Implements guardrails that intercept agent outputs and tool calls, applying content filtering, prompt injection detection, and rate limiting. Uses a policy engine to define allowed actions per agent, with audit logging of all capability requests and denials.","intents":["Prevent agents from accessing files or APIs they shouldn't have permission to use","Detect and block prompt injection attempts before they reach the LLM","Audit all agent actions for compliance and security investigation","Define fine-grained permissions per agent (e.g., read-only file access, specific API endpoints)"],"best_for":["Organizations deploying untrusted or third-party agents in production","Teams handling sensitive data requiring strict access control and audit trails","Developers building multi-tenant agent platforms with per-tenant isolation","Security-conscious teams needing compliance-ready agent deployments"],"limitations":["Capability declarations are developer-provided; no automatic capability inference from agent code","Prompt injection detection uses heuristics; sophisticated attacks may bypass detection","Audit logging adds ~20-50ms latency per agent action due to synchronous policy evaluation","No built-in capability revocation; requires restart to update agent permissions"],"requires":["TypeScript runtime (Node.js 18+)","Policy configuration file defining agent capabilities","Persistent storage for audit logs (file system or database)"],"input_types":["agent capability declarations","policy definitions (JSON or YAML)","agent action requests (tool calls, API requests)"],"output_types":["allow/deny decisions","audit log entries","capability violation reports"],"categories":["safety-moderation","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cowork-os--cowork-os__cap_3","uri":"capability://automation.workflow.self.hosted.deployment.with.local.first.architecture","name":"self-hosted deployment with local-first architecture","description":"Enables fully self-hosted deployment where CoWork-OS runs on user infrastructure (macOS desktop, Linux server, or Docker container) without requiring cloud services for core agent execution. Supports local LLM inference via Ollama integration, local message storage, and optional cloud provider integration (Claude, GPT) only when explicitly configured. Uses Electron for desktop deployment on macOS with native system integrations (iMessage, file system access), and Docker for server deployments.","intents":["Run AI agents entirely on-premises without sending data to third-party cloud services","Use local LLM models (via Ollama) for privacy-sensitive tasks","Deploy agents on personal hardware (Mac, Linux server) with full control","Integrate with local system resources (files, clipboard, native APIs) without cloud intermediaries"],"best_for":["Organizations with strict data residency or privacy requirements","Developers building personal AI assistants with local-first privacy","Teams wanting to avoid cloud vendor lock-in for agent infrastructure","Users with sensitive data who cannot use cloud LLM providers"],"limitations":["Local LLM inference (Ollama) significantly slower than cloud providers; expect 2-10x latency increase","Requires substantial local compute resources for running LLMs; minimum 8GB RAM, GPU recommended","No built-in high-availability or failover; single-machine deployments have no redundancy","Desktop (Electron) deployment limited to macOS; Linux/Windows support requires Docker or manual setup","Scaling beyond single machine requires manual orchestration; no built-in clustering"],"requires":["macOS 10.15+ for Electron desktop app, or Linux/Windows with Docker","Node.js 18+ runtime","Ollama installed locally for local LLM inference (optional but recommended for privacy)","8GB+ RAM minimum, GPU recommended for local LLM performance","Docker and Docker Compose for containerized deployments"],"input_types":["agent configuration files","local file system paths","system environment variables"],"output_types":["local agent logs","file system artifacts","local message history database"],"categories":["automation-workflow","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cowork-os--cowork-os__cap_4","uri":"capability://tool.use.integration.model.context.protocol.mcp.server.implementation.and.client.integration","name":"model context protocol (mcp) server implementation and client integration","description":"Implements MCP as both server (exposing agent capabilities as MCP resources and tools) and client (consuming MCP servers from other systems). Agents can declare tools and resources following MCP specification, allowing external systems to discover and invoke agent capabilities through standardized MCP protocol. Supports MCP server spawning, lifecycle management, and bidirectional communication with proper error handling and timeout management.","intents":["Expose agent capabilities as MCP resources so external systems can discover and use them","Integrate with other MCP-compatible systems and tools without custom API wrappers","Allow agents to consume MCP servers from other applications (e.g., Claude Desktop MCP servers)","Build composable agent ecosystems where agents can call other agents' capabilities via MCP"],"best_for":["Developers building MCP-compatible agent ecosystems","Teams integrating CoWork-OS agents with Claude Desktop or other MCP clients","Organizations wanting standardized agent-to-agent communication","Builders creating composable multi-agent systems"],"limitations":["MCP protocol overhead adds ~50-200ms per server call due to JSON-RPC serialization","MCP server lifecycle management (spawn, monitor, restart) adds operational complexity","No built-in MCP server discovery; servers must be manually configured","Timeout handling is basic; long-running MCP operations may timeout without retry logic","MCP specification is still evolving; breaking changes possible in future versions"],"requires":["TypeScript runtime (Node.js 18+)","MCP server binaries or Node.js packages for servers to be integrated","Configuration file defining MCP servers and their endpoints"],"input_types":["MCP resource requests (JSON-RPC format)","tool invocation requests with parameters","MCP server configuration"],"output_types":["MCP resource responses","tool execution results","MCP protocol messages"],"categories":["tool-use-integration","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cowork-os--cowork-os__cap_5","uri":"capability://memory.knowledge.persistent.conversation.state.management.with.context.window.optimization","name":"persistent conversation state management with context window optimization","description":"Manages multi-turn conversation history with automatic context window optimization that summarizes or truncates old messages to fit within LLM token limits while preserving conversation semantics. Stores conversation state locally (or in configured database) with per-user and per-channel isolation. Implements sliding window strategy where recent messages are kept verbatim, older messages are summarized, and very old messages are archived, with configurable retention policies.","intents":["Maintain conversation context across multiple turns without losing important information","Automatically manage token budget to fit conversations within LLM context windows","Retrieve conversation history for auditing, debugging, or user review","Isolate conversations per user and per channel to prevent context leakage"],"best_for":["Multi-turn agent conversations requiring long-term context","Teams needing conversation audit trails and history","Developers building agents with limited context windows (e.g., using smaller local models)","Applications where conversation continuity across sessions is critical"],"limitations":["Summarization of old messages may lose nuanced context; quality depends on summarization model","Context window optimization adds ~100-300ms per turn due to summarization and truncation logic","No built-in vector search for semantic retrieval; finding relevant old messages requires linear scan","Storage grows linearly with conversation length; no automatic cleanup without explicit policies","Conversation isolation is logical only; requires careful implementation to prevent cross-user leakage"],"requires":["TypeScript runtime (Node.js 18+)","Local file storage or database (SQLite, PostgreSQL) for conversation history","LLM provider for summarization (can use same provider as agent or separate)"],"input_types":["new user messages","conversation history (array of message objects)","context window size limit"],"output_types":["optimized message history (within token limit)","conversation metadata (user ID, channel, timestamps)","summarization artifacts"],"categories":["memory-knowledge","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cowork-os--cowork-os__cap_6","uri":"capability://automation.workflow.agent.configuration.and.orchestration.with.yaml.json.policy.files","name":"agent configuration and orchestration with yaml/json policy files","description":"Defines agents through declarative configuration files (YAML/JSON) specifying agent name, system prompt, available tools, capability requirements, provider preferences, and channel assignments. Orchestrates agent lifecycle (initialization, message routing, tool execution) based on configuration without requiring code changes. Supports agent composition where one agent can delegate to other agents based on task routing rules defined in configuration.","intents":["Define and deploy agents without writing code for each agent variant","Switch agent behavior (prompts, tools, providers) by editing configuration files","Route different user intents to different specialized agents based on configuration rules","Manage multiple agents with different capabilities and access controls from single deployment"],"best_for":["Non-technical users configuring agents through YAML/JSON","Teams deploying multiple agent variants without code duplication","Organizations needing to update agent behavior without redeployment","Developers building agent platforms where customers define agents via config"],"limitations":["Complex agent logic (conditional branching, dynamic tool selection) becomes unwieldy in config files; requires code for sophisticated behavior","Configuration validation is basic; invalid configs may fail at runtime rather than parse time","No built-in hot-reload; agent configuration changes require restart","Agent composition (delegating between agents) adds latency and complexity; no built-in optimization for call chains"],"requires":["TypeScript runtime (Node.js 18+)","YAML or JSON configuration files following CoWork-OS schema","Schema validation tool (optional but recommended)"],"input_types":["YAML/JSON configuration files","system prompts (text)","tool definitions (JSON schema)"],"output_types":["initialized agent instances","routing decisions (which agent handles request)","configuration validation errors"],"categories":["automation-workflow","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cowork-os--cowork-os__cap_7","uri":"capability://tool.use.integration.tool.calling.with.schema.based.function.registry.and.provider.native.bindings","name":"tool calling with schema-based function registry and provider-native bindings","description":"Implements tool calling through a schema-based function registry where tools are defined as JSON schemas with implementation functions. Translates tool schemas to provider-native formats (OpenAI function calling, Anthropic tool_use, Gemini function calling, Ollama tool calling) and handles tool execution with automatic parameter validation and error handling. Supports both synchronous and asynchronous tool implementations with timeout management.","intents":["Define tools once and use them across multiple LLM providers without provider-specific code","Validate tool parameters against schema before execution to catch errors early","Execute tools with timeout protection to prevent hanging agent processes","Chain multiple tool calls in sequence with automatic result passing between tools"],"best_for":["Developers building agents with multiple tools across different LLM providers","Teams needing strict parameter validation for tool execution","Applications requiring timeout protection for external tool calls","Builders creating tool marketplaces or plugin systems"],"limitations":["Schema-based validation adds ~10-30ms per tool call due to JSON schema validation","Provider-native tool calling has different semantics (e.g., OpenAI vs. Anthropic); abstraction may hide important differences","Async tool execution requires careful error handling; timeouts may leave resources in inconsistent state","Tool result formatting varies by provider; abstraction may lose provider-specific metadata","No built-in tool versioning; updating tool schemas requires careful migration"],"requires":["TypeScript runtime (Node.js 18+)","Tool implementations (functions or async functions)","JSON schema definitions for each tool"],"input_types":["JSON schema tool definitions","tool implementation functions","tool invocation requests from LLM"],"output_types":["tool execution results","parameter validation errors","timeout errors"],"categories":["tool-use-integration","code-generation-editing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cowork-os--cowork-os__cap_8","uri":"capability://safety.moderation.prompt.injection.detection.and.content.filtering.with.configurable.rules","name":"prompt injection detection and content filtering with configurable rules","description":"Applies content filtering and prompt injection detection to both user inputs and agent outputs through configurable rule sets. Uses pattern matching, keyword detection, and heuristic analysis to identify potential injection attempts (e.g., 'ignore previous instructions', 'system prompt override'). Filters outputs for prohibited content (profanity, PII, sensitive data) based on configurable word lists and regex patterns. Supports custom filter implementations for domain-specific requirements.","intents":["Prevent users from injecting prompts that override agent instructions","Filter agent outputs to remove sensitive data (PII, API keys, internal information)","Enforce content policies (no profanity, no hate speech) on both inputs and outputs","Audit filtered content for security investigation and policy violation tracking"],"best_for":["Teams deploying agents in untrusted environments (public chatbots, user-facing applications)","Organizations with strict content policies or compliance requirements","Developers building multi-tenant agent platforms with per-tenant content policies","Security-conscious teams needing defense-in-depth against prompt injection"],"limitations":["Pattern-based detection is brittle; sophisticated injection attempts (obfuscation, encoding) may bypass detection","False positives common with keyword-based filtering; legitimate content may be incorrectly filtered","Filtering adds ~20-50ms latency per message due to pattern matching","Custom filter implementations require domain expertise; no automatic policy generation","No machine learning-based detection; relies on heuristics rather than learned patterns"],"requires":["TypeScript runtime (Node.js 18+)","Filter rule configuration (regex patterns, keyword lists)","Optional: custom filter implementation functions"],"input_types":["user input text","agent output text","filter rule definitions (regex, keywords)"],"output_types":["filtered text (with injections/prohibited content removed)","filter violation reports","audit logs"],"categories":["safety-moderation","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"github-cowork-os--cowork-os__cap_9","uri":"capability://automation.workflow.rate.limiting.and.quota.management.per.agent.user.and.channel","name":"rate limiting and quota management per agent, user, and channel","description":"Enforces rate limits and quotas at multiple levels: per-agent (total requests), per-user (requests per user), and per-channel (requests per channel). Implements token bucket algorithm for smooth rate limiting with configurable burst allowances. Tracks usage across time windows (per minute, per hour, per day) and enforces hard limits with graceful degradation (queuing, rejection, or fallback). Integrates with LLM provider quotas to prevent exceeding provider limits.","intents":["Prevent single user or agent from consuming all resources and starving other users","Stay within LLM provider rate limits and quotas without hitting hard limits","Implement fair usage policies with different quotas for different user tiers","Monitor and alert on quota usage approaching limits"],"best_for":["Multi-user agent platforms with shared resources","Teams managing costs by enforcing per-user or per-agent quotas","Applications with strict SLA requirements needing fair resource allocation","Developers building agent marketplaces with per-tenant quotas"],"limitations":["Rate limiting adds ~5-10ms latency per request due to quota checking","Distributed rate limiting (across multiple servers) requires shared state store (Redis); local-only implementation doesn't scale","Quota enforcement is hard limit; no built-in graceful degradation (e.g., reduced quality vs. rejection)","No built-in cost tracking; quota management is request-count based, not cost-based","Quota reset timing (daily, weekly) requires careful clock synchronization in distributed systems"],"requires":["TypeScript runtime (Node.js 18+)","Quota configuration (limits per agent/user/channel)","Optional: Redis or similar for distributed rate limiting"],"input_types":["request metadata (user ID, agent ID, channel)","quota configuration (limits, time windows)","usage tracking data"],"output_types":["allow/deny decisions","remaining quota information","quota violation alerts"],"categories":["automation-workflow","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":42,"verified":false,"data_access_risk":"high","permissions":["TypeScript runtime (Node.js 18+)","Platform API credentials (Twilio account for WhatsApp, Telegram Bot Token, Discord Bot Token, Slack Bot Token)","macOS system for iMessage channel support","Network connectivity to platform APIs or self-hosted message broker","API keys for desired providers (OpenAI, Anthropic, Google, or local Ollama instance)","Network connectivity to cloud providers or local Ollama server","macOS 10.15+","Electron runtime (included in CoWork-OS desktop app)","User permission grants for file system and clipboard access","Persistent storage for logs (file system, database, or log aggregation service)"],"failure_modes":["Platform-specific features (rich media, interactive buttons) require custom adapter code per channel","Rate limiting and quota management must be handled per-platform; no unified rate limiter","Message ordering guarantees depend on underlying platform reliability; no built-in deduplication","iMessage support limited to macOS; requires running on Apple hardware for that channel","Model-specific capabilities (vision, function calling) require provider-specific code paths despite abstraction","Prompt engineering optimizations for one model may not transfer to another; abstraction hides model-specific tuning","Cost tracking and quota management per provider must be implemented separately","Streaming response handling adds ~50-100ms latency due to abstraction layer translation","Desktop integration limited to macOS; no native Windows or Linux support (Docker alternative available)","File access permissions are coarse-grained; no per-file or per-directory granularity","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.30497871996690507,"quality":0.49,"ecosystem":0.6000000000000001,"match_graph":0.25,"freshness":0.75,"weights":{"adoption":0.25,"quality":0.25,"ecosystem":0.1,"match_graph":0.28,"freshness":0.12}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:21.549Z","last_scraped_at":"2026-05-03T13:59:57.743Z","last_commit":"2026-05-03T13:47:12Z"},"community":{"stars":269,"forks":46,"weekly_downloads":null,"model_downloads":null,"model_likes":null}},"distribution":{"claim_url":"https://unfragile.ai/submit?claim=cowork-os--cowork-os","compare_url":"https://unfragile.ai/compare?artifact=cowork-os--cowork-os"}},"signature":"R1y7LITwPwjX6KKMrmtHkCtBhejD4A14DePfxuylYFhFO5bPwq06/Egn+O9PfsZ6UFREAcLfJuxX3Z5UAcLBBw==","signedAt":"2026-06-22T15:29:52.330Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/cowork-os--cowork-os","artifact":"https://unfragile.ai/cowork-os--cowork-os","verify":"https://unfragile.ai/api/v1/verify?slug=cowork-os--cowork-os","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"}}