{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"tool_yuna","slug":"yuna","name":"Yuna","type":"product","url":"https://www.yuna.io","page_url":"https://unfragile.ai/yuna","categories":["chatbots-assistants"],"tags":[],"pricing":{"model":"freemium","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"tool_yuna__cap_0","uri":"capability://text.generation.language.conversational.cbt.coaching.via.voice.and.text.chat","name":"conversational cbt coaching via voice and text chat","description":"Delivers real-time cognitive behavioral therapy techniques through a dual-modal interface (voice transcription + text chat), processing user input through an unspecified LLM to generate contextually-aware therapeutic responses. The system maintains conversation state across sessions to reference prior mood patterns and therapeutic progress, enabling continuity without human therapist involvement. Responses are framed around CBT principles (thought-behavior-emotion linkage, cognitive restructuring) but implementation mechanism (prompt engineering vs. fine-tuning vs. structured outputs) is undocumented.","intents":["I need mental health support outside therapy hours without scheduling friction","I want to practice CBT techniques interactively when anxiety or depression symptoms emerge","I need accessible mental health coaching that doesn't require affording a therapist"],"best_for":["individuals with mild-to-moderate anxiety or depression seeking supplementary support","employees in organizations offering Yuna as an EAP benefit","users in regions with limited therapist availability or high therapy costs"],"limitations":["Cannot diagnose mental health conditions or provide clinical assessment","No capability to prescribe medication or manage pharmacological treatment","Cannot handle acute crisis situations—only escalates to external resources","Conversation context window size unknown; coherence over extended multi-day conversations unverified","LLM model identity undisclosed; no transparency on training data or fine-tuning approach","Response latency and voice transcription quality metrics not published"],"requires":["Internet connection for cloud-based SaaS access","Microphone and speaker for voice mode (or text-only alternative)","User account creation (authentication mechanism unspecified)","Acceptance of data collection for mood tracking and aggregation"],"input_types":["voice (codec and language support unspecified)","text chat (character limits unknown)","structured mood self-report (likely Likert scale or similar)"],"output_types":["conversational text or synthesized voice response","CBT-framed therapeutic guidance","escalation triggers with crisis resource links"],"categories":["text-generation-language","mental-health-support"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_yuna__cap_1","uri":"capability://safety.moderation.real.time.escalation.detection.with.crisis.resource.routing","name":"real-time escalation detection with crisis resource routing","description":"Monitors conversation content in real-time to identify crisis indicators (suicidal ideation, severe self-harm, acute psychosis) and automatically triggers escalation workflows that surface crisis resources (hotline links, emergency contacts) to the user. Detection mechanism is undocumented but likely uses keyword matching, sentiment analysis, or LLM-based classification against a crisis taxonomy. Upon escalation trigger, system initiates proactive check-in messaging and routes alert data to HR dashboard (if deployed in enterprise context) while maintaining claimed privacy boundary that individual conversation content is not exposed to HR.","intents":["I need the system to catch when a user is in crisis and provide immediate resources","I want to ensure users in severe distress are not left without intervention options","I need to track escalation frequency for compliance and risk management purposes"],"best_for":["HR teams deploying mental health benefits and needing liability protection","organizations in regulated industries (healthcare, finance) requiring crisis response documentation","enterprises seeking to reduce EAP costs while maintaining duty-of-care obligations"],"limitations":["Detection methodology (keyword, ML classifier, LLM-based) is undocumented; false positive and false negative rates unknown","Cannot provide live crisis intervention—only routes to external resources (hotlines, emergency services)","No capability to contact emergency services directly or override user agency","Crisis resource database and update frequency unknown","Escalation trigger criteria not published; users cannot understand what language/behavior triggers alerts","No mention of multi-language crisis resource localization despite 155-country deployment claim"],"requires":["Continuous conversation monitoring (implies persistent connection or polling)","Crisis resource database (maintained by Yuna, update SLA unknown)","Integration with HR dashboard for escalation logging (enterprise deployments only)","User consent to escalation detection and resource sharing"],"input_types":["conversation text or voice transcription","mood tracking data (optional context for escalation scoring)"],"output_types":["escalation flag with severity level (unknown schema)","crisis resource links (specific resources and format unknown)","proactive check-in message to user","escalation event logged to HR dashboard (anonymized)"],"categories":["safety-moderation","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_yuna__cap_10","uri":"capability://text.generation.language.voice.interface.with.transcription.and.synthesis","name":"voice interface with transcription and synthesis","description":"Supports voice input (speech-to-text transcription) and voice output (text-to-speech synthesis) as alternatives to text chat, enabling hands-free conversational interaction. Voice interface is positioned as accessibility feature and natural interaction modality, but specific implementation details are undocumented: transcription service provider (Google, AWS, Azure, proprietary?), supported languages, accent handling, latency, and synthesis quality are all unknown. Voice capability is mentioned as core feature but lacks technical depth.","intents":["I want to talk to the AI naturally without typing, especially when in distress","I need accessibility support for visual impairment or typing difficulty","I want to use mental health support while driving or doing other activities"],"best_for":["users with visual impairment or motor disabilities requiring voice interface","individuals preferring natural conversation over text input","people using Yuna while multitasking (driving, exercising, etc.)"],"limitations":["Transcription service provider is undocumented; unclear if using Google, AWS, Azure, or proprietary service","Supported languages for voice are undocumented; unclear if voice supports all 155 countries or subset","Transcription accuracy and accent handling are undocumented","Voice synthesis quality and latency are undocumented; unclear if synthesis is natural-sounding or robotic","No mention of voice authentication or speaker identification","Voice input codec and compression are undocumented","No mention of noise cancellation or audio quality requirements","Fallback behavior for voice failures (transcription errors, synthesis failures) is undocumented"],"requires":["Microphone and speaker hardware","Internet connection for transcription and synthesis services","Voice-capable device (smartphone, computer, smart speaker, etc.)","Supported language for voice transcription and synthesis"],"input_types":["voice audio (codec unknown)","speech in supported language"],"output_types":["text transcription of user speech","synthesized voice response","text alternative (if voice synthesis fails)"],"categories":["text-generation-language","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_yuna__cap_11","uri":"capability://safety.moderation.clinical.expert.involvement.in.system.design.and.content.curation","name":"clinical expert involvement in system design and content curation","description":"Claims system is 'built by clinical experts' and uses 'evidence-backed' therapeutic techniques, suggesting involvement of mental health professionals in system design, content curation, and validation. However, specific clinical expertise (psychiatrists? psychologists? therapists?), involvement scope (design review? content creation? ongoing validation?), and evidence base (published research? clinical trials? expert consensus?) are entirely undocumented. This claim is positioned as differentiation but lacks verifiable substance.","intents":["I want assurance that mental health support is clinically sound, not just AI-generated","I need to know that therapeutic content is reviewed by qualified mental health professionals","I want evidence that AI-delivered CBT is effective for my condition"],"best_for":["users skeptical of AI-delivered mental health support seeking clinical credibility","HR teams needing to justify mental health benefits to clinical advisory boards","organizations in regulated industries requiring clinical validation"],"limitations":["Clinical expertise type is undocumented; unclear if psychiatrists, psychologists, therapists, or other professionals are involved","Involvement scope is undocumented; unclear if clinical experts designed system, curated content, validated outputs, or provided advisory input","Evidence base is undocumented; no published research, clinical trials, or expert consensus cited","No mention of ongoing clinical validation or quality assurance processes","No mention of clinical advisory board or external review","No mention of adverse event reporting or clinical safety monitoring","Clinical credentials of involved experts are not disclosed","No mention of clinical guidelines (APA, NICE, etc.) used to inform system design"],"requires":["Clinical expert involvement (type and scope unknown)","Evidence base for therapeutic techniques (unknown)","Clinical validation methodology (unknown)"],"input_types":["clinical expertise and guidance"],"output_types":["clinically-informed system design","evidence-backed therapeutic content","clinical validation reports (if any)"],"categories":["safety-moderation","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_yuna__cap_2","uri":"capability://data.processing.analysis.daily.mood.tracking.with.historical.pattern.aggregation","name":"daily mood tracking with historical pattern aggregation","description":"Collects structured user self-reports of mood (likely via Likert scale or similar) on a daily cadence, stores mood data points with timestamps, and aggregates historical patterns to feed into subsequent conversation context and HR analytics dashboards. The system uses mood data to personalize therapeutic responses (e.g., recognizing deteriorating trends) and to populate real-time HR dashboards with team-level well-being metrics ('no surveys required' implies sentiment extraction from conversations, though mechanism is undocumented). Mood data is claimed to be anonymized before HR exposure, but individual-to-aggregate mapping is not transparent.","intents":["I want to track my mental health trends over time without manual survey burden","I need HR to understand team well-being without invasive individual surveys","I want the AI to reference my mood history to provide contextually-aware support"],"best_for":["HR teams seeking real-time well-being metrics without survey fatigue","individuals wanting passive mood tracking integrated into therapy workflow","organizations needing compliance documentation of employee mental health interventions"],"limitations":["Mood data collection mechanism (check-in frequency, scale type, optional vs. mandatory) is undocumented","Data retention policy unknown; unclear how long mood history is preserved","No user control over data deletion or export of mood history","Aggregation methodology for HR dashboard is undocumented; potential for re-identification of individuals from aggregate trends","No mention of mood data validation (e.g., detecting random/spam responses)","Cross-session mood context feeding into LLM is undocumented; unclear if mood data is included in conversation context window"],"requires":["Daily user engagement (check-in frequency unknown)","Structured mood input capability (UI component for Likert scale or similar)","Backend time-series database for mood storage and aggregation","User consent to mood data collection and HR dashboard exposure"],"input_types":["structured mood self-report (scale type unknown)","optional timestamp metadata","conversation content (for sentiment extraction, if applicable)"],"output_types":["mood data point with timestamp","aggregated mood trend (daily, weekly, monthly)","team-level well-being metrics for HR dashboard","mood context for conversational AI (if fed into LLM)"],"categories":["data-processing-analysis","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_yuna__cap_3","uri":"capability://data.processing.analysis.hr.dashboard.with.real.time.aggregated.team.well.being.metrics","name":"hr dashboard with real-time aggregated team well-being metrics","description":"Provides HR teams with real-time visualization of anonymized, aggregated well-being metrics derived from employee interactions with Yuna (usage frequency, engagement trends, team-level mood patterns, escalation event counts). The dashboard is designed to surface organizational mental health trends without exposing individual conversation content or identifiable user data, enabling HR to justify mental health benefit ROI and identify at-risk teams. Aggregation logic and anonymization methodology are undocumented; unclear how individual data is de-identified and whether re-identification is possible through trend analysis.","intents":["I need to demonstrate ROI of mental health benefits to executive leadership","I want to identify teams with declining well-being trends for proactive intervention","I need compliance documentation showing mental health support utilization"],"best_for":["HR leaders in mid-to-large enterprises (100+ employees) deploying Yuna as EAP","organizations in regulated industries requiring mental health benefit documentation","companies seeking data-driven justification for mental health spending"],"limitations":["Aggregation granularity unknown (team-level? department-level? company-level?)","Metrics included in dashboard are undocumented (usage count, engagement score, mood average, escalation frequency?)","Anonymization methodology is undocumented; no transparency on de-identification techniques or re-identification risk","Real-time latency unknown; 'real-time' claim unverified against actual dashboard update frequency","No mention of custom metric definition or drill-down capability into aggregate trends","Data retention policy for dashboard metrics unknown","No API access mentioned; dashboard is UI-only, limiting integration with HRIS or BI tools"],"requires":["Enterprise deployment of Yuna with minimum employee count (threshold unknown)","HR admin account with dashboard access permissions","Aggregation backend infrastructure (Yuna-managed, no on-premise option mentioned)","Employee consent to data aggregation and HR dashboard exposure"],"input_types":["aggregated interaction data from all users","mood tracking data (aggregated)","escalation event logs (anonymized)"],"output_types":["real-time dashboard visualizations (chart types unknown)","aggregated metrics (specific metrics unknown)","trend reports (format and frequency unknown)","exportable reports (format unknown)"],"categories":["data-processing-analysis","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_yuna__cap_4","uri":"capability://text.generation.language.dbt.inspired.skills.coaching.with.evidence.backed.session.structure","name":"dbt-inspired skills coaching with evidence-backed session structure","description":"Delivers structured coaching sessions focused on dialectical behavior therapy (DBT) skills (distress tolerance, emotion regulation, mindfulness, interpersonal effectiveness) through conversational interaction. Sessions are described as 'short' and 'evidence-backed' but implementation details are undocumented: unclear whether sessions follow a fixed curriculum, whether skills are sequenced based on user needs, or whether the LLM generates DBT content dynamically vs. retrieving from a curated skill library. Coaching is positioned as supplementary to CBT (primary modality) rather than a replacement for DBT therapy.","intents":["I want to learn and practice DBT skills (emotion regulation, distress tolerance) in a structured format","I need short, focused coaching sessions on specific DBT techniques when I'm in distress","I want evidence-based skills training without committing to full DBT therapy"],"best_for":["individuals with emotion dysregulation or distress tolerance deficits seeking skill-building","users already in therapy who want supplementary DBT practice between sessions","people with borderline personality disorder or complex trauma seeking accessible DBT exposure"],"limitations":["DBT skills curriculum structure is undocumented; unclear if skills are sequenced or user-selected","No mention of homework assignments, practice tracking, or skill mastery assessment","Session length and structure unknown; 'short' is subjective and unverified","No mention of DBT-specific assessment (e.g., distress tolerance or emotion regulation baseline)","Cannot provide individual DBT therapy (which requires trained therapist and multi-modal treatment)","No mention of skills coaching being personalized to user's specific deficits","Evidence base for AI-delivered DBT coaching is not cited or verified"],"requires":["User request or system recommendation to initiate skills coaching session","Conversational interface (voice or text)","Potential prerequisite: baseline understanding of DBT concepts (not documented)"],"input_types":["user request for specific skill (e.g., 'teach me emotion regulation')","conversational interaction during coaching session","optional: user self-report of current distress level"],"output_types":["structured coaching content on DBT skill","interactive practice exercises or examples","guidance on applying skill in real-world context","optional: homework or follow-up suggestions"],"categories":["text-generation-language","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_yuna__cap_5","uri":"capability://text.generation.language.multi.language.conversational.support.across.155.countries","name":"multi-language conversational support across 155 countries","description":"Claims to deliver conversational mental health support across 155 countries, implying multi-language capability, but specific supported languages are undocumented. Language support likely includes voice transcription, text chat, and response generation in multiple languages, but localization of CBT/DBT content, crisis resources, and therapeutic framing across cultural contexts is not mentioned. No information on language detection, fallback behavior for unsupported languages, or translation quality assurance.","intents":["I need mental health support in my native language, not English","I want crisis resources and therapeutic guidance localized to my country's healthcare system","I need the system to automatically detect and respond in my preferred language"],"best_for":["multinational enterprises with geographically distributed workforces","individuals in non-English-speaking countries seeking accessible mental health support","organizations expanding mental health benefits to international offices"],"limitations":["Supported languages are not documented; unclear which of 155 countries have language support","No mention of language detection or user language preference setting","Crisis resources are likely US-centric (HIPAA-aligned suggests US focus); localization to country-specific hotlines/services unknown","CBT/DBT content may not be culturally adapted; therapeutic framing may not align with non-Western mental health concepts","Voice transcription quality and accent handling unknown","No mention of language-specific fine-tuning or cultural sensitivity review","Fallback behavior for unsupported languages unknown (English default? Error message?)"],"requires":["Language support for user's preferred language (specific languages unknown)","Voice capability for transcription in supported languages (codec and quality unknown)","Localized crisis resource database (if applicable to user's country)"],"input_types":["voice in supported language","text chat in supported language","optional: explicit language preference setting"],"output_types":["conversational response in user's language","localized crisis resources (if available)","translated CBT/DBT content (quality unknown)"],"categories":["text-generation-language","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_yuna__cap_6","uri":"capability://automation.workflow.proactive.check.in.messaging.with.escalation.follow.up","name":"proactive check-in messaging with escalation follow-up","description":"Automatically initiates outreach messages to users following escalation detection, checking in on their well-being and reinforcing crisis resources. Check-in messaging is triggered by escalation events and likely uses templated or LLM-generated messages, but frequency, timing, and personalization logic are undocumented. This capability bridges the gap between passive escalation (resource routing) and active intervention (human follow-up), positioning Yuna as a semi-active safety system rather than purely reactive.","intents":["I want the system to follow up with users after crisis detection to ensure they're safe","I need to reduce the likelihood of users ignoring crisis resources by reinforcing outreach","I want to create a sense of continuity and care even when human intervention isn't possible"],"best_for":["HR teams deploying Yuna as part of comprehensive mental health safety strategy","organizations seeking to demonstrate duty-of-care through documented follow-up","enterprises wanting to reduce liability exposure from crisis escalations"],"limitations":["Check-in message frequency and timing are undocumented; unclear if check-ins are one-time or recurring","Message content is undocumented; unclear if messages are templated, personalized, or LLM-generated","No mention of user opt-out capability for check-in messages","Escalation follow-up success metrics are unknown (e.g., did user engage with resources?)","No mention of escalation severity classification; unclear if all escalations trigger check-ins or only high-severity events","Check-in message delivery mechanism unknown (in-app notification? SMS? Email?)","No mention of human review or override of automated check-ins"],"requires":["Escalation detection trigger (see escalation detection capability)","User contact information (email, phone, or in-app notification capability)","Message delivery infrastructure (notification service, SMS gateway, email service)","Escalation event logging and follow-up tracking"],"input_types":["escalation event with severity level (unknown schema)","user contact preferences (unknown format)"],"output_types":["check-in message (content and format unknown)","delivery confirmation (unknown logging)","user engagement tracking (unknown metrics)"],"categories":["automation-workflow","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_yuna__cap_7","uri":"capability://safety.moderation.hipaa.aligned.and.soc.2.type.ii.certified.cloud.infrastructure","name":"hipaa-aligned and soc 2 type ii certified cloud infrastructure","description":"Operates on a cloud-based SaaS infrastructure with claimed HIPAA compliance (for US healthcare data) and SOC 2 Type II certification (for security, availability, processing integrity controls). This indicates managed cloud deployment (likely AWS, Azure, or GCP) with encryption, access controls, audit logging, and third-party security assessment. However, specific infrastructure details (multi-tenant vs. single-tenant, data residency options, backup/disaster recovery SLAs) are undocumented, and compliance scope (GDPR, CCPA, PIPEDA for non-US regions) is unclear.","intents":["I need to ensure user mental health data is protected under HIPAA regulations","I want third-party verification that Yuna's infrastructure meets security standards","I need to deploy mental health support in regulated industries (healthcare, finance) with compliance confidence"],"best_for":["US-based enterprises in regulated industries (healthcare, finance, government) requiring HIPAA compliance","organizations with security-conscious procurement teams requiring SOC 2 certification","companies handling sensitive employee health data needing compliance documentation"],"limitations":["HIPAA compliance scope is undocumented; unclear if all data is HIPAA-protected or only certain data types","SOC 2 Type II certification scope is undocumented; unclear which controls are assessed (security, availability, processing integrity, confidentiality, privacy?)","No mention of GDPR, CCPA, PIPEDA, or other non-US privacy regulations despite 155-country deployment claim","Data residency options unknown; unclear if data can be stored in specific regions (EU, APAC, etc.)","Backup and disaster recovery SLAs are undocumented","Encryption methodology (in-transit, at-rest, key management) is undocumented","No mention of data breach notification procedures or incident response SLAs","Multi-tenant vs. single-tenant architecture unknown; potential for cross-tenant data leakage if multi-tenant"],"requires":["Cloud infrastructure provider (AWS, Azure, GCP, or proprietary)","HIPAA Business Associate Agreement (BAA) for US healthcare deployments","SOC 2 Type II audit and certification (third-party assessed)","Encryption and access control infrastructure"],"input_types":["user mental health data (conversations, mood tracking, escalation events)"],"output_types":["encrypted data storage","audit logs","compliance reports (HIPAA, SOC 2)"],"categories":["safety-moderation","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_yuna__cap_8","uri":"capability://automation.workflow.freemium.pricing.model.with.unspecified.tier.structure","name":"freemium pricing model with unspecified tier structure","description":"Operates on a freemium model (free tier + paid premium) to lower barrier to entry for cost-conscious users, but specific tier features, pricing, and paywall locations are undocumented. Website contains no pricing page or free trial details, making it impossible to assess feature differentiation, per-user cost, or upgrade triggers. Freemium positioning suggests free access to core conversational support with premium features (likely enhanced analytics, priority support, or advanced coaching) behind paywall, but this is inferred rather than documented.","intents":["I want to try mental health support without upfront cost commitment","I need to understand what features are free vs. paid before deploying to my organization","I want transparent pricing to justify mental health benefit ROI to leadership"],"best_for":["individuals exploring mental health support options with limited budget","HR teams evaluating Yuna before enterprise deployment","cost-conscious organizations seeking affordable mental health benefits"],"limitations":["Pricing page is not accessible or documented; no pricing information available","Free tier features are undocumented; unclear what conversational support is free vs. paid","Premium tier features are undocumented; unclear what features justify upgrade","Per-user pricing model unknown; unclear if pricing is per-employee, per-conversation, or per-feature","Volume discounts or enterprise pricing unknown","Free trial duration (if any) is undocumented","Upgrade triggers and paywall locations are undocumented","No mention of annual vs. monthly billing or contract terms"],"requires":["User account creation (free tier)","Payment method for premium tier (if applicable)","Minimum seat requirements for enterprise deployment (unknown)"],"input_types":["user account information"],"output_types":["free tier access (features unknown)","premium tier access (features unknown)","pricing quote (for enterprise deployments)"],"categories":["automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"tool_yuna__cap_9","uri":"capability://memory.knowledge.conversation.history.persistence.with.cross.session.context","name":"conversation history persistence with cross-session context","description":"Maintains conversation history across user sessions, enabling the system to reference prior interactions, mood trends, and therapeutic progress in subsequent conversations. This persistence mechanism (likely a vector database or conversation store) allows the LLM to maintain therapeutic continuity without requiring users to re-explain their situation. However, context window management, history retention policy, and how historical context is fed into the LLM are undocumented, creating uncertainty about conversation coherence over extended time periods.","intents":["I want the AI to remember my previous conversations and therapeutic progress","I need continuity in support without re-explaining my situation each time","I want the system to reference mood trends and past coping strategies I've discussed"],"best_for":["individuals using Yuna regularly (daily or multiple times per week) for ongoing support","users with chronic mental health conditions requiring consistent therapeutic context","people seeking to build therapeutic relationship with AI over time"],"limitations":["Context window size is unknown; unclear how much historical context is included in each LLM call","History retention policy is undocumented; unclear how long conversations are stored (days? months? indefinitely?)","No mention of user control over history deletion or privacy","Context prioritization logic is undocumented; unclear if recent conversations are weighted more heavily than old ones","No mention of conversation summarization; unclear if full history is stored or summarized for efficiency","Cross-session context feeding into LLM is undocumented; unclear if history is retrieved via semantic search, recency ranking, or other mechanism","No mention of conversation export or user data portability"],"requires":["Persistent user account with conversation storage","Backend database or vector store for conversation history","Context retrieval mechanism (semantic search, recency ranking, etc.)","LLM context window large enough to include historical context"],"input_types":["current user input (text or voice)","historical conversation data (retrieved from storage)"],"output_types":["conversational response with historical context awareness","mood trend references","therapeutic progress acknowledgment"],"categories":["memory-knowledge","text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":40,"verified":false,"data_access_risk":"high","permissions":["Internet connection for cloud-based SaaS access","Microphone and speaker for voice mode (or text-only alternative)","User account creation (authentication mechanism unspecified)","Acceptance of data collection for mood tracking and aggregation","Continuous conversation monitoring (implies persistent connection or polling)","Crisis resource database (maintained by Yuna, update SLA unknown)","Integration with HR dashboard for escalation logging (enterprise deployments only)","User consent to escalation detection and resource sharing","Microphone and speaker hardware","Internet connection for transcription and synthesis services"],"failure_modes":["Cannot diagnose mental health conditions or provide clinical assessment","No capability to prescribe medication or manage pharmacological treatment","Cannot handle acute crisis situations—only escalates to external resources","Conversation context window size unknown; coherence over extended multi-day conversations unverified","LLM model identity undisclosed; no transparency on training data or fine-tuning approach","Response latency and voice transcription quality metrics not published","Detection methodology (keyword, ML classifier, LLM-based) is undocumented; false positive and false negative rates unknown","Cannot provide live crisis intervention—only routes to external resources (hotlines, emergency services)","No capability to contact emergency services directly or override user agency","Crisis resource database and update frequency unknown","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.31666666666666665,"quality":0.72,"ecosystem":0.15000000000000002,"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":"2026-05-24T12:16:34.117Z","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=yuna","compare_url":"https://unfragile.ai/compare?artifact=yuna"}},"signature":"q8Lv0HLNgKzTo3SIyBUfLM0i+ukdUzgoEEm4XS1bTc7Gglur+d8smbm8G3dxwyOP82Q5NJ2TMPLPfcuh5xYYCQ==","signedAt":"2026-06-22T14:34:45.119Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/yuna","artifact":"https://unfragile.ai/yuna","verify":"https://unfragile.ai/api/v1/verify?slug=yuna","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"}}