{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"tool_cxcortex","slug":"cxcortex","name":"CXCortex","type":"product","url":"https://www.hostcomm.co.uk","page_url":"https://unfragile.ai/cxcortex","categories":["chatbots-assistants"],"tags":[],"pricing":{"model":"freemium","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"tool_cxcortex__cap_0","uri":"capability://data.processing.analysis.real.time.customer.interaction.analytics.and.insight.extraction","name":"real-time customer interaction analytics and insight extraction","description":"Processes incoming customer interaction data (calls, chats, emails, tickets) through a streaming analytics pipeline that identifies patterns, sentiment, intent, and resolution outcomes in real-time without batch processing delays. The system appears to use event-driven architecture to capture interaction metadata and apply NLP-based classification to surface actionable insights immediately, enabling support teams to spot trends and quality issues as they occur rather than in post-shift reports.","intents":["I need to understand what customers are asking about right now, not in tomorrow's report","I want to identify support quality issues and agent performance gaps in real-time so I can coach or escalate immediately","I need to surface emerging customer pain points before they become widespread complaints"],"best_for":["mid-market customer support teams with 10-100 agents","customer research teams analyzing interaction patterns","operations managers optimizing support workflows"],"limitations":["Real-time processing latency likely increases with interaction volume; freemium tier may have throughput caps (e.g., max 100 concurrent interactions)","Accuracy of intent/sentiment classification depends on training data quality and may struggle with domain-specific jargon or non-English languages","Requires continuous data streaming; batch-processed historical data may not benefit from real-time insights"],"requires":["Integration with at least one communication channel (Zendesk, Intercom, Twilio, or native API)","Minimum interaction volume to train classification models (likely 1000+ labeled interactions)","API credentials for connected systems"],"input_types":["text (chat, email, ticket transcripts)","call transcripts or audio metadata","structured interaction metadata (duration, agent ID, customer segment)"],"output_types":["structured analytics dashboards (sentiment scores, intent categories, resolution rates)","real-time alerts (high-priority issues, quality flags)","aggregated reports (trends, patterns, agent performance metrics)"],"categories":["data-processing-analysis","customer-intelligence"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_cxcortex__cap_1","uri":"capability://planning.reasoning.personalized.customer.interaction.recommendations.and.next.best.action","name":"personalized customer interaction recommendations and next-best-action","description":"Analyzes customer history, behavior, preferences, and interaction context to generate personalized recommendations for support agents or automated systems on how to handle each interaction. The system likely maintains a customer profile graph (interaction history, purchase behavior, sentiment trajectory, previous resolutions) and uses collaborative filtering or contextual bandit algorithms to suggest the highest-probability resolution path or communication approach for each customer segment.","intents":["I want my support agents to know the best way to handle each customer based on their history and preferences","I need to suggest the right product, offer, or resolution to each customer in real-time during their interaction","I want to reduce resolution time by recommending the most likely successful next step for each interaction type"],"best_for":["customer support teams handling diverse customer segments with varying preferences","sales-enabled support teams (smarketing) needing to recommend upsells or cross-sells","retention teams targeting at-risk customers with personalized interventions"],"limitations":["Personalization quality depends on data richness; new customers or those with sparse interaction history receive generic recommendations","Freemium tier likely limits personalization depth (e.g., segment-level recommendations only, not individual-level)","Requires historical data to train recommendation models; cold-start problem for new customer segments or product lines","Privacy constraints (GDPR, CCPA) may limit how deeply customer profiles can be built or shared across teams"],"requires":["Customer data platform or CRM integration (Salesforce, HubSpot, Pipedrive) with historical interaction logs","Minimum 3-6 months of interaction history per customer segment to train models","Defined business outcomes (resolution time, CSAT, upsell rate) to optimize recommendations against"],"input_types":["customer profile data (demographics, purchase history, lifetime value)","interaction history (previous tickets, chat transcripts, call notes)","real-time interaction context (current issue, customer sentiment, channel)"],"output_types":["ranked recommendations (next-best-action, suggested resolution, recommended offer)","confidence scores and reasoning (why this recommendation)","A/B test results (which recommendations drive better outcomes)"],"categories":["planning-reasoning","personalization"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_cxcortex__cap_10","uri":"capability://data.processing.analysis.customer.sentiment.tracking.and.emotional.intelligence.scoring","name":"customer sentiment tracking and emotional intelligence scoring","description":"Analyzes customer sentiment and emotional tone throughout interactions using NLP-based emotion detection, tracking sentiment changes over time and across interactions to identify at-risk or highly satisfied customers. The system likely uses transformer-based models (BERT, RoBERTa) to classify emotions (frustration, satisfaction, urgency) from text and generates alerts when sentiment drops significantly or customer frustration escalates.","intents":["I want to identify frustrated customers early so I can intervene before they churn","I need to track customer satisfaction trends over time to understand if we're improving or declining","I want to flag interactions where customer frustration is escalating so agents can adjust their approach"],"best_for":["customer success or retention teams focused on churn prevention","support teams wanting to improve customer satisfaction proactively","organizations with high customer lifetime value where retention is critical"],"limitations":["Sentiment analysis accuracy varies by language, domain, and tone; sarcasm, cultural differences, and complex emotions may be misclassified","Freemium tier likely limits sentiment tracking frequency or historical depth (e.g., 30 days only, basic sentiment only)","Sentiment alone doesn't predict churn; requires correlation with behavioral signals (support tickets, usage, payment issues)","Privacy concerns if sentiment data is used for automated decision-making (e.g., auto-escalating frustrated customers)"],"requires":["Integration with communication channels to access interaction text","Interaction transcripts or text data for sentiment analysis","Historical sentiment data to establish baselines and trends"],"input_types":["interaction text (chat, email, call transcripts)","customer metadata (history, account status)"],"output_types":["sentiment scores (positive, negative, neutral, with confidence)","emotion classifications (frustration, satisfaction, urgency, etc.)","sentiment trends (over time, by customer, by issue type)","alerts (sentiment drop, escalating frustration)"],"categories":["data-processing-analysis","emotional-intelligence"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_cxcortex__cap_2","uri":"capability://automation.workflow.automated.task.routing.and.workflow.orchestration","name":"automated task routing and workflow orchestration","description":"Automatically routes incoming customer interactions (tickets, chats, calls) to the most appropriate agent, team, or automated system based on issue classification, agent availability, skill matching, and workload balancing. The system likely implements a rule engine or ML-based routing model that evaluates multiple routing criteria (priority, complexity, agent expertise, current queue depth) and orchestrates handoffs between human agents and automated systems (chatbots, knowledge base, escalation workflows).","intents":["I want to automatically route customer issues to the right agent without manual assignment or queue management","I need to balance workload across my support team so no one is overloaded while others are idle","I want to escalate complex issues to specialists and handle simple issues with automation to reduce manual overhead"],"best_for":["support teams with 20+ agents across multiple skill levels or specializations","organizations with high interaction volume (100+ daily interactions) where manual routing is inefficient","teams using a mix of human agents and chatbots/automation"],"limitations":["Routing accuracy depends on issue classification quality; misclassified issues may be routed to wrong agent, increasing resolution time","Workload balancing may not account for issue complexity or agent fatigue; simple queue-depth balancing can lead to uneven effort distribution","Freemium tier likely limits routing rules or agent skill tags, forcing basic round-robin or priority-only routing","Requires continuous tuning; routing rules that work for one team composition may fail after hiring or turnover"],"requires":["Integration with ticketing system (Zendesk, Jira Service Desk, Freshdesk) or communication platform (Intercom, Slack)","Agent skill tags or competency matrix defined in system","Minimum 2-4 weeks of historical routing data to train ML models (if using ML-based routing)"],"input_types":["incoming interaction metadata (issue type, priority, customer segment, channel)","agent availability and skill inventory (current queue, expertise tags, availability status)","historical routing outcomes (resolution time, customer satisfaction, escalation rate)"],"output_types":["routing decision (assigned agent, team, or automation path)","routing confidence and reasoning","queue status and wait time estimates"],"categories":["automation-workflow","task-routing"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_cxcortex__cap_3","uri":"capability://automation.workflow.automated.task.execution.and.administrative.workflow.automation","name":"automated task execution and administrative workflow automation","description":"Automates repetitive administrative tasks (ticket creation, status updates, customer notifications, knowledge base updates, follow-up scheduling) by executing predefined workflows triggered by interaction events or time-based rules. The system likely uses a workflow engine (state machine or DAG-based) that chains together API calls to connected systems (CRM, ticketing, email, Slack) to reduce manual data entry and context-switching for support teams.","intents":["I want to automatically create tickets, update statuses, and send notifications without manual data entry","I need to schedule follow-ups and reminders for unresolved issues without relying on agent memory","I want to reduce the time my team spends on administrative work so they can focus on customer conversations"],"best_for":["support teams with high administrative overhead (>20% of time spent on non-customer-facing tasks)","organizations with complex multi-step workflows (e.g., escalation → investigation → customer update → closure)","teams using multiple disconnected tools (CRM, ticketing, email, Slack) that require manual data sync"],"limitations":["Workflow automation is brittle; changes to downstream systems (API changes, field renames) can break workflows without alerting","Freemium tier likely limits number of automations or workflow complexity (e.g., max 5 active workflows, no nested conditions)","Error handling and retry logic may not be transparent; failed automations may silently fail or require manual intervention","Requires careful design to avoid creating duplicate tickets or sending redundant notifications"],"requires":["Integration with at least 2-3 downstream systems (CRM, ticketing, email, Slack, etc.)","API credentials and permissions for all connected systems","Defined workflow logic (trigger conditions, action sequences, error handling)"],"input_types":["interaction events (ticket created, chat ended, call completed)","time-based triggers (daily, weekly, on-demand)","conditional logic (if sentiment is negative, if resolution time > threshold, etc.)"],"output_types":["executed actions (tickets created, statuses updated, notifications sent)","workflow execution logs (success/failure, timestamps, affected records)","audit trail for compliance"],"categories":["automation-workflow","task-automation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_cxcortex__cap_4","uri":"capability://data.processing.analysis.customer.interaction.data.aggregation.and.unified.view","name":"customer interaction data aggregation and unified view","description":"Aggregates customer interaction data from multiple channels (email, chat, phone, social media, tickets) into a unified customer profile or interaction timeline, enabling support agents to see complete customer history without switching between systems. The system likely implements a data lake or unified API layer that normalizes interaction data from disparate sources and maintains a single source of truth for customer context.","intents":["I want my support agents to see all customer interactions in one place instead of jumping between email, chat, and ticketing systems","I need to understand the full customer journey across channels to provide better context and personalized support","I want to avoid asking customers to repeat information they've already shared in previous interactions"],"best_for":["organizations using 3+ communication channels (email, chat, phone, social, tickets)","support teams with high context-switching overhead","customer-centric organizations prioritizing seamless omnichannel experience"],"limitations":["Data aggregation latency may introduce stale data; real-time sync across all channels is difficult and resource-intensive","Freemium tier likely limits number of integrated channels or data retention period (e.g., 30 days of history only)","Data normalization is lossy; channel-specific metadata (call duration, email attachments, chat sentiment) may be lost in unified view","Privacy and compliance complexity increases with more data sources; GDPR/CCPA requirements may limit data retention or sharing"],"requires":["Integration with 2+ communication platforms (Zendesk, Intercom, Twilio, Gmail, Slack, etc.)","API credentials and permissions for all connected systems","Data warehouse or lake infrastructure to store and query unified data"],"input_types":["interaction data from multiple channels (emails, chat transcripts, call logs, tickets, social messages)","customer metadata (CRM records, purchase history, account status)"],"output_types":["unified customer profile (interaction timeline, contact info, account details)","channel-specific views (email thread, chat history, call transcript)","aggregated metrics (total interactions, satisfaction, lifetime value)"],"categories":["data-processing-analysis","integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_cxcortex__cap_5","uri":"capability://data.processing.analysis.customer.satisfaction.and.quality.scoring.with.automated.feedback.collection","name":"customer satisfaction and quality scoring with automated feedback collection","description":"Automatically collects customer satisfaction feedback (CSAT, NPS, CES) through post-interaction surveys or sentiment analysis of interaction transcripts, and scores interaction quality based on predefined criteria (resolution, politeness, first-contact resolution). The system likely uses NLP to extract sentiment from text and combines survey responses with behavioral signals (repeat contacts, escalations) to generate a holistic quality score for each interaction and agent.","intents":["I want to measure customer satisfaction without manually sending surveys or waiting for responses","I need to identify low-quality interactions or problematic agents so I can coach or retrain them","I want to correlate satisfaction with support metrics (resolution time, agent expertise) to understand what drives customer happiness"],"best_for":["support teams with quality assurance programs or performance management needs","organizations prioritizing customer satisfaction metrics (CSAT, NPS)","teams wanting to reduce survey fatigue by using automated sentiment analysis instead of manual surveys"],"limitations":["Sentiment analysis accuracy varies by language, domain, and tone; sarcasm or complex emotions may be misclassified","Survey response rates may be low (typically 5-15%), biasing results toward highly satisfied or dissatisfied customers","Freemium tier likely limits survey frequency or quality scoring depth (e.g., basic CSAT only, no detailed quality rubrics)","Correlation between satisfaction and support metrics is correlational, not causal; requires careful interpretation"],"requires":["Integration with communication channels to collect feedback (email, SMS, in-app surveys)","Interaction transcripts or text data for sentiment analysis","Defined quality criteria or rubric for scoring (if using rule-based scoring)"],"input_types":["post-interaction surveys (CSAT, NPS, CES questions)","interaction transcripts (chat, email, call transcripts)","behavioral signals (repeat contacts, escalations, resolution status)"],"output_types":["satisfaction scores (CSAT %, NPS, CES)","quality scores (per interaction, per agent, per team)","trend reports (satisfaction over time, by agent, by issue type)","alerts (low satisfaction, quality issues)"],"categories":["data-processing-analysis","quality-assurance"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_cxcortex__cap_6","uri":"capability://search.retrieval.knowledge.base.integration.and.ai.powered.answer.suggestion","name":"knowledge base integration and ai-powered answer suggestion","description":"Integrates with internal knowledge bases (Confluence, SharePoint, custom wikis) and uses semantic search or retrieval-augmented generation (RAG) to suggest relevant articles or answers to support agents or customers during interactions. The system likely embeds knowledge base articles into a vector database and uses similarity search to find relevant content based on customer questions, reducing agent research time and enabling self-service for customers.","intents":["I want my support agents to quickly find relevant knowledge base articles without manually searching","I want to enable customers to self-serve by suggesting relevant articles before they contact support","I need to reduce resolution time by providing agents with pre-populated answers from our knowledge base"],"best_for":["support teams with large knowledge bases (100+ articles) covering common issues","organizations wanting to reduce support volume through self-service","teams with high first-contact resolution targets"],"limitations":["Knowledge base quality directly impacts suggestion quality; outdated or poorly written articles reduce usefulness","Semantic search may miss relevant articles if knowledge base uses different terminology than customer questions","Freemium tier likely limits knowledge base size or search frequency (e.g., max 500 articles, 100 searches/day)","Requires continuous knowledge base maintenance; orphaned or conflicting articles degrade suggestion quality"],"requires":["Integration with knowledge base platform (Confluence, SharePoint, Zendesk, custom API)","Minimum 50-100 well-structured articles covering common issues","Vector database or embedding service (e.g., OpenAI embeddings, Pinecone, Weaviate)"],"input_types":["customer questions or issue descriptions (text)","knowledge base articles (text, HTML, Markdown)"],"output_types":["ranked article suggestions (title, URL, relevance score)","extracted answer snippets (relevant passages from articles)","self-service recommendations (for customer-facing portals)"],"categories":["search-retrieval","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_cxcortex__cap_7","uri":"capability://data.processing.analysis.agent.performance.analytics.and.coaching.insights","name":"agent performance analytics and coaching insights","description":"Tracks agent performance metrics (response time, resolution time, customer satisfaction, first-contact resolution rate, adherence to scripts) and generates coaching insights or performance comparisons to help managers identify improvement opportunities. The system likely aggregates interaction-level metrics into agent-level dashboards and uses statistical analysis or peer benchmarking to highlight outliers or trends requiring attention.","intents":["I want to see which agents are performing well and which need coaching or training","I need to identify specific performance gaps (e.g., slow response time, low satisfaction) and recommend targeted coaching","I want to benchmark agent performance against team averages to set realistic improvement targets"],"best_for":["support managers with 5-50 direct reports","organizations with performance management or quality assurance programs","teams wanting data-driven coaching instead of subjective evaluations"],"limitations":["Metrics can be gamed; agents may prioritize response time over quality, or cherry-pick easy issues","Freemium tier likely limits metrics depth or historical data retention (e.g., 30 days only, basic metrics only)","Correlation between individual metrics and business outcomes (retention, revenue) is not always clear","Privacy concerns if performance data is too granular or shared too broadly; requires careful access controls"],"requires":["Integration with ticketing/communication systems to extract interaction-level metrics","Minimum 2-4 weeks of interaction data per agent to establish baselines","Defined performance targets or benchmarks"],"input_types":["interaction-level data (response time, resolution time, customer satisfaction, issue type)","agent metadata (tenure, skill tags, team assignment)","historical performance data for benchmarking"],"output_types":["agent performance dashboards (individual and team-level metrics)","performance trends (improvement/decline over time)","peer benchmarking (vs team average, vs top performers)","coaching recommendations (specific areas for improvement)"],"categories":["data-processing-analysis","performance-management"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_cxcortex__cap_8","uri":"capability://tool.use.integration.multi.channel.customer.communication.orchestration","name":"multi-channel customer communication orchestration","description":"Enables support teams to manage customer conversations across multiple channels (email, chat, phone, SMS, social media) from a unified interface, allowing agents to respond to customers on their preferred channel without context loss. The system likely implements a channel abstraction layer that normalizes message formats and maintains conversation state across channels, enabling seamless handoffs between channels if needed.","intents":["I want my agents to handle customer conversations across all channels without switching between different tools","I need to maintain conversation context when customers switch channels (e.g., start in chat, continue in email)","I want to route customers to their preferred communication channel instead of forcing them to use our preferred channel"],"best_for":["organizations with customers across multiple communication preferences (email, chat, phone, SMS)","support teams wanting to reduce tool switching and context loss","customer-centric organizations prioritizing channel choice"],"limitations":["Channel integration complexity increases with each new channel; some channels (social media, SMS) have unique constraints (character limits, threading)","Freemium tier likely limits number of integrated channels or message volume (e.g., max 3 channels, 1000 messages/month)","Conversation state management is complex; context may be lost if customer switches channels mid-conversation","Compliance requirements vary by channel (SMS regulations, social media archiving); requires careful implementation"],"requires":["Integration with 2+ communication platforms (email, Slack, Twilio, Facebook, Twitter, etc.)","API credentials and permissions for all connected channels","Unified inbox or agent interface supporting all channels"],"input_types":["messages from multiple channels (email, chat, SMS, social)","customer metadata (channel preferences, contact info)"],"output_types":["unified inbox view (all messages across channels)","channel-specific message formatting (respecting character limits, threading)","conversation history (across channels)"],"categories":["tool-use-integration","omnichannel"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_cxcortex__cap_9","uri":"capability://planning.reasoning.predictive.issue.escalation.and.priority.routing","name":"predictive issue escalation and priority routing","description":"Predicts which customer issues are likely to escalate or require specialist intervention based on issue characteristics, customer history, and interaction patterns, and automatically routes high-risk issues to senior agents or escalation queues. The system likely uses classification models trained on historical escalation data to identify early warning signals (customer sentiment, issue complexity, customer value) and trigger proactive escalation before issues become critical.","intents":["I want to identify issues that will likely escalate before they become critical","I need to route complex or high-value customer issues to specialists instead of junior agents","I want to reduce escalation rates by catching issues early and routing them appropriately"],"best_for":["support teams with high escalation rates (>15%) or long resolution times","organizations with tiered support (junior agents, specialists, escalation teams)","customer-centric organizations prioritizing high-value customer retention"],"limitations":["Escalation prediction is probabilistic; false positives (over-escalating simple issues) waste specialist time, while false negatives (missing escalation-prone issues) increase resolution time","Freemium tier likely limits escalation rules or prediction frequency (e.g., basic rules only, no ML-based prediction)","Requires historical escalation data to train models; new issue types or customer segments may not be predicted accurately","Escalation routing may create bottlenecks if too many issues are routed to specialists"],"requires":["Integration with ticketing system to extract escalation history","Minimum 3-6 months of historical data with escalation outcomes","Defined escalation criteria or specialist skill tags"],"input_types":["issue characteristics (type, priority, complexity)","customer data (lifetime value, history, sentiment)","interaction context (channel, agent skill level, current queue)"],"output_types":["escalation probability scores (0-100%)","recommended routing (specialist team, escalation queue)","reasoning (which factors triggered escalation prediction)"],"categories":["planning-reasoning","predictive-analytics"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":43,"verified":false,"data_access_risk":"high","permissions":["Integration with at least one communication channel (Zendesk, Intercom, Twilio, or native API)","Minimum interaction volume to train classification models (likely 1000+ labeled interactions)","API credentials for connected systems","Customer data platform or CRM integration (Salesforce, HubSpot, Pipedrive) with historical interaction logs","Minimum 3-6 months of interaction history per customer segment to train models","Defined business outcomes (resolution time, CSAT, upsell rate) to optimize recommendations against","Integration with communication channels to access interaction text","Interaction transcripts or text data for sentiment analysis","Historical sentiment data to establish baselines and trends","Integration with ticketing system (Zendesk, Jira Service Desk, Freshdesk) or communication platform (Intercom, Slack)"],"failure_modes":["Real-time processing latency likely increases with interaction volume; freemium tier may have throughput caps (e.g., max 100 concurrent interactions)","Accuracy of intent/sentiment classification depends on training data quality and may struggle with domain-specific jargon or non-English languages","Requires continuous data streaming; batch-processed historical data may not benefit from real-time insights","Personalization quality depends on data richness; new customers or those with sparse interaction history receive generic recommendations","Freemium tier likely limits personalization depth (e.g., segment-level recommendations only, not individual-level)","Requires historical data to train recommendation models; cold-start problem for new customer segments or product lines","Privacy constraints (GDPR, CCPA) may limit how deeply customer profiles can be built or shared across teams","Sentiment analysis accuracy varies by language, domain, and tone; sarcasm, cultural differences, and complex emotions may be misclassified","Freemium tier likely limits sentiment tracking frequency or historical depth (e.g., 30 days only, basic sentiment only)","Sentiment alone doesn't predict churn; requires correlation with behavioral signals (support tickets, usage, payment issues)","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.36666666666666664,"quality":0.78,"ecosystem":0.2,"match_graph":0.25,"freshness":0.75,"weights":{"adoption":0.25,"quality":0.25,"ecosystem":0.1,"match_graph":0.35,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:30.282Z","last_scraped_at":"2026-04-05T13:23:42.552Z","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=cxcortex","compare_url":"https://unfragile.ai/compare?artifact=cxcortex"}},"signature":"sI1WnuNCyIvVAjTdT7rvzEsUdX9I4Gjf1bJ35ONneK8GPeHpzEIbW2Skr362rujpPxyliOK2Uq1pZ+OGLTxUAw==","signedAt":"2026-06-21T12:50:06.437Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/cxcortex","artifact":"https://unfragile.ai/cxcortex","verify":"https://unfragile.ai/api/v1/verify?slug=cxcortex","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"}}