{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"awesome-avanzai","slug":"avanzai","name":"Avanzai","type":"agent","url":"https://avanz.ai/","page_url":"https://unfragile.ai/avanzai","categories":["ai-agents"],"tags":[],"pricing":{"model":"unknown","free":false,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"awesome-avanzai__cap_0","uri":"capability://planning.reasoning.multi.asset.portfolio.risk.quantification.via.agent.reasoning","name":"multi-asset portfolio risk quantification via agent reasoning","description":"Decomposes portfolio risk assessment into discrete agent tasks that analyze correlations, volatility, and tail risks across equities, fixed income, commodities, and alternatives. Uses agentic reasoning loops to iteratively refine risk estimates by querying market data APIs, computing Value-at-Risk (VaR) and Expected Shortfall (ES) metrics, and synthesizing results into actionable risk profiles. The agent maintains context across multiple asset classes and time horizons to produce holistic portfolio risk scores.","intents":["Quantify total portfolio risk exposure across 10+ asset classes without manual spreadsheet consolidation","Identify concentration risks and tail-event scenarios that traditional static models miss","Generate risk reports that explain which positions drive portfolio volatility and drawdown potential"],"best_for":["Institutional asset managers and hedge funds automating daily risk reporting","Wealth advisors managing multi-asset client portfolios who need real-time risk dashboards","Risk officers at banks needing continuous portfolio monitoring across trading desks"],"limitations":["Agentic reasoning adds latency (typically 5-30 seconds per portfolio analysis) compared to pre-computed risk models","Accuracy depends on quality and timeliness of underlying market data feeds — stale prices degrade risk estimates","Backtesting and stress-testing capabilities unknown; may not capture regime-change risks in extreme market conditions","No explicit handling of illiquid assets or private equity positions that lack continuous pricing"],"requires":["Real-time or near-real-time market data feed (Bloomberg, Reuters, or alternative data provider)","Portfolio holdings data in structured format (CSV, JSON, or direct broker API connection)","API credentials for Avanzai platform","Minimum portfolio size/complexity threshold unknown"],"input_types":["structured portfolio data (holdings, quantities, entry prices)","market prices and volatility surfaces","correlation matrices or time-series price data","optional: custom risk parameters (confidence levels, time horizons)"],"output_types":["risk metrics (VaR, ES, Sharpe ratio, maximum drawdown)","risk attribution by position and asset class","natural language risk summaries and alerts","structured JSON risk profiles for downstream systems"],"categories":["planning-reasoning","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-avanzai__cap_1","uri":"capability://planning.reasoning.dynamic.asset.allocation.optimization.with.constraint.satisfaction","name":"dynamic asset allocation optimization with constraint satisfaction","description":"Orchestrates multi-objective optimization agents that rebalance portfolios subject to regulatory constraints, tax efficiency targets, and liquidity requirements. The system uses constraint-satisfaction reasoning to navigate competing objectives (maximize return, minimize risk, minimize tax drag, respect position limits) and generates rebalancing recommendations with execution sequencing. Agents evaluate trade-offs between objectives and surface Pareto-optimal allocation frontiers to decision-makers.","intents":["Automatically rebalance portfolios to target allocations while minimizing tax impact and transaction costs","Generate compliant allocation recommendations that respect regulatory position limits and concentration rules","Explore trade-offs between return, risk, and tax efficiency to find allocations aligned with client preferences"],"best_for":["Robo-advisors and digital wealth platforms automating portfolio rebalancing for thousands of clients","Institutional asset managers optimizing allocations across multiple mandates with different constraints","Tax-aware portfolio managers seeking to minimize tax leakage while maintaining strategic allocations"],"limitations":["Constraint satisfaction complexity grows exponentially with number of positions and constraints — may timeout on very large portfolios (1000+ holdings)","Tax-loss harvesting logic depends on accurate cost-basis tracking; errors in historical data propagate to recommendations","No explicit handling of behavioral constraints (e.g., client aversion to selling winners) — requires manual override capability","Rebalancing frequency and thresholds must be pre-configured; no adaptive logic for market regime changes"],"requires":["Current portfolio holdings with cost basis and acquisition dates","Target allocation policy or risk profile specification","Tax rate assumptions and tax-loss harvesting rules","Liquidity constraints and position limit rules (regulatory or internal)","Real-time or daily pricing for all holdings"],"input_types":["portfolio composition (holdings, quantities, cost basis)","target allocation weights or risk profile","constraint specifications (position limits, sector caps, regulatory rules)","tax parameters (marginal tax rate, holding periods, wash-sale rules)","market prices and transaction costs"],"output_types":["rebalancing recommendations (buy/sell quantities and sequencing)","projected tax impact and after-tax returns","allocation efficiency metrics (Sharpe ratio, tax-adjusted return)","Pareto frontier of allocation trade-offs","execution instructions for order management systems"],"categories":["planning-reasoning","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-avanzai__cap_2","uri":"capability://automation.workflow.real.time.portfolio.monitoring.with.anomaly.detection.and.alerts","name":"real-time portfolio monitoring with anomaly detection and alerts","description":"Deploys continuous monitoring agents that track portfolio metrics (returns, volatility, correlations, drawdowns) against baselines and thresholds, detecting deviations that signal risk or opportunity. Uses statistical anomaly detection (z-score, isolation forest, or learned baselines) to distinguish signal from noise and triggers escalating alerts (email, SMS, dashboard) when thresholds are breached. Agents maintain rolling windows of historical metrics to adapt baselines to market regime changes.","intents":["Receive real-time alerts when portfolio volatility, drawdown, or correlation structure deviates from expected ranges","Detect unexpected position movements or market dislocations that may require immediate action","Monitor compliance with risk limits and regulatory thresholds without manual daily review"],"best_for":["Active traders and hedge fund managers needing sub-minute alerts on portfolio stress","Risk officers monitoring compliance with internal risk limits across multiple portfolios","Wealth advisors wanting to proactively flag client portfolios at risk of breaching investment policy statements"],"limitations":["Alert fatigue risk if thresholds are not carefully tuned — false positives reduce actionability","Anomaly detection baselines require sufficient historical data (typically 60+ days) to be reliable; new portfolios may generate spurious alerts","No causal analysis — alerts flag deviations but don't explain root causes (e.g., single position move vs. market-wide shock)","Latency depends on data feed frequency; daily pricing limits alert timeliness for intraday traders"],"requires":["Real-time or near-real-time portfolio pricing (intraday for active traders, daily for passive investors)","Historical portfolio data (minimum 60 days for baseline calibration)","Configurable alert thresholds and escalation rules","Notification infrastructure (email, SMS, webhook, or dashboard integration)"],"input_types":["streaming portfolio prices and returns","historical portfolio metrics (returns, volatility, correlations)","threshold and alert configuration","optional: market regime indicators or macro data for context"],"output_types":["real-time alerts (email, SMS, webhook, dashboard notification)","anomaly scores and severity levels","historical metric charts and trend analysis","contextual information (e.g., which positions drove the anomaly)"],"categories":["automation-workflow","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-avanzai__cap_3","uri":"capability://planning.reasoning.scenario.analysis.and.stress.testing.via.agent.simulation","name":"scenario analysis and stress testing via agent simulation","description":"Orchestrates agent-driven scenario analysis that simulates portfolio behavior under hypothetical market conditions (interest rate shocks, equity crashes, volatility spikes, geopolitical events). Agents parameterize scenarios, apply shock vectors to market prices and correlations, recompute portfolio metrics, and synthesize results into scenario reports. Uses Monte Carlo simulation or historical scenario replay to generate distributions of outcomes rather than point estimates.","intents":["Understand how portfolio would perform under historical crises (2008 financial crisis, COVID crash, etc.) or hypothetical stress scenarios","Identify portfolio vulnerabilities to specific risk factors (interest rates, credit spreads, commodity prices)","Generate scenario reports for client communication or regulatory stress-testing requirements"],"best_for":["Institutional asset managers and pension funds required to conduct regulatory stress tests","Risk managers at banks and insurance companies assessing tail risks and capital adequacy","Wealth advisors wanting to educate clients on downside scenarios and portfolio resilience"],"limitations":["Scenario design is subjective — results depend heavily on assumptions about shock magnitudes and correlation changes","Historical scenarios may not capture unprecedented events; Monte Carlo simulations require distributional assumptions that may not hold in tail events","Computational cost scales with portfolio complexity and number of scenarios — large portfolios may require hours to run full stress tests","No feedback loop to optimize portfolio structure based on scenario results; recommendations require manual interpretation"],"requires":["Historical price data for all portfolio holdings (minimum 5-10 years for Monte Carlo calibration)","Scenario definitions (shock vectors, correlation changes, or historical event parameters)","Computational resources for Monte Carlo simulation (GPU acceleration optional but recommended)","Market data for scenario application (current prices, volatilities, correlations)"],"input_types":["portfolio composition and current prices","scenario specifications (historical events, hypothetical shocks, or Monte Carlo parameters)","historical price data for calibration","optional: custom correlation matrices or factor models"],"output_types":["scenario portfolio returns and risk metrics","distribution of outcomes (mean, percentiles, tail metrics)","position-level impact analysis (which holdings drive scenario losses)","scenario reports with visualizations and narrative summaries"],"categories":["planning-reasoning","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-avanzai__cap_4","uri":"capability://planning.reasoning.multi.agent.portfolio.collaboration.and.consensus.building","name":"multi-agent portfolio collaboration and consensus building","description":"Coordinates multiple specialized agents (risk agent, return agent, tax agent, compliance agent) that evaluate portfolios from different perspectives and reach consensus on recommendations. Agents debate trade-offs, surface conflicts (e.g., tax efficiency vs. risk reduction), and synthesize recommendations that balance competing objectives. Uses negotiation or voting protocols to resolve disagreements and produce final recommendations with transparency on trade-offs.","intents":["Generate portfolio recommendations that balance multiple objectives (risk, return, tax efficiency, compliance) without requiring manual trade-off specification","Understand conflicts between different portfolio management goals and make informed trade-off decisions","Audit portfolio decisions by reviewing how different agents evaluated the portfolio and what trade-offs were made"],"best_for":["Large institutional asset managers with multiple portfolio management teams needing coordinated recommendations","Compliance-heavy organizations (banks, insurance) needing transparent audit trails of portfolio decisions","Multi-objective portfolio optimization where trade-offs are complex and require human judgment"],"limitations":["Agent consensus mechanisms add latency and complexity — may be slower than single-objective optimization","Disagreement resolution depends on voting/negotiation protocol; different protocols may produce different recommendations","Transparency on trade-offs is valuable but may overwhelm non-technical stakeholders with too much detail","Requires careful tuning of agent objectives and weights to avoid biased consensus toward particular goals"],"requires":["Specification of agent roles and objectives (risk minimization, return maximization, tax efficiency, compliance)","Consensus protocol (voting, negotiation, weighted averaging, or custom logic)","Portfolio data and market data for all agents to evaluate","Audit logging infrastructure to track agent reasoning and decisions"],"input_types":["portfolio composition and market data","agent objectives and constraints","consensus protocol specification","optional: historical decisions or preferences for training consensus models"],"output_types":["consensus portfolio recommendations","agent-level evaluations and disagreements","trade-off analysis (e.g., tax cost of risk reduction)","audit trail of agent reasoning and consensus process"],"categories":["planning-reasoning","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-avanzai__cap_5","uri":"capability://text.generation.language.natural.language.portfolio.explanation.and.reporting","name":"natural language portfolio explanation and reporting","description":"Generates natural language summaries and reports that explain portfolio composition, risk metrics, allocation changes, and recommendations in plain English. Uses templated generation with agent reasoning to select relevant metrics, highlight key insights, and tailor explanations to audience (technical vs. non-technical). Integrates with portfolio data and metrics to produce dynamic reports that update as portfolio changes.","intents":["Generate client-facing portfolio reports that explain allocation, performance, and risk in non-technical language","Create internal memos explaining portfolio decisions and trade-offs to stakeholders","Produce regulatory reports (e.g., risk disclosures) with consistent, auditable language"],"best_for":["Wealth advisors and robo-advisors needing to communicate portfolio strategy to retail clients","Institutional asset managers producing quarterly or annual reports for limited partners","Compliance teams generating regulatory disclosures and risk summaries"],"limitations":["Natural language generation quality depends on training data and templates — may produce generic or inaccurate summaries if not carefully tuned","No guarantee of factual accuracy — generated text should be reviewed by humans before client distribution","Tone and style customization requires additional configuration; one-size-fits-all templates may not suit all audiences","Regulatory language must be carefully reviewed for compliance; generated text may not meet legal requirements without human review"],"requires":["Portfolio data and metrics (returns, risk, allocation, changes)","Templates or examples of desired report format and tone","Audience specification (retail client, institutional investor, regulator)","Optional: brand guidelines or style preferences"],"input_types":["portfolio metrics and performance data","allocation changes and rebalancing history","risk analysis and scenario results","optional: market commentary or macro context"],"output_types":["natural language portfolio summaries and reports","client-facing explanations of allocation and risk","regulatory disclosures and risk summaries","formatted reports (PDF, HTML, email-ready text)"],"categories":["text-generation-language"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-avanzai__cap_6","uri":"capability://tool.use.integration.integration.with.external.data.sources.and.market.feeds","name":"integration with external data sources and market feeds","description":"Provides agent-driven connectors to external market data providers (Bloomberg, Reuters, Yahoo Finance, alternative data vendors) and portfolio systems (custodians, brokers, trading platforms). Agents handle authentication, data transformation, and reconciliation across sources, normalizing heterogeneous data formats into unified portfolio and market data models. Supports both batch ingestion and streaming real-time data feeds.","intents":["Ingest portfolio data from multiple custodians and brokers without manual consolidation","Access real-time market prices and volatility data from multiple providers with automatic fallback and reconciliation","Integrate with existing portfolio management systems and workflows without replacing them"],"best_for":["Multi-custodian asset managers needing unified portfolio view across multiple brokers","Organizations with legacy portfolio systems wanting to add AI-driven analytics without rip-and-replace","Firms using multiple market data vendors and needing unified, reconciled pricing"],"limitations":["Data quality and timeliness depend on external providers — stale or incorrect data from sources degrades analysis","Authentication and API management complexity increases with number of data sources; requires careful credential and rate-limit management","Data transformation and reconciliation logic must be maintained as external APIs change; breaking changes in provider APIs require code updates","Latency depends on external provider performance; slow data feeds bottleneck real-time analysis"],"requires":["API credentials for external data providers (Bloomberg, Reuters, custodians, etc.)","Network connectivity and firewall rules allowing outbound API calls","Data schema specifications for portfolio and market data models","Optional: dedicated data infrastructure (message queues, data lakes) for high-volume streaming data"],"input_types":["API credentials and connection parameters","data source specifications (which providers, which data types)","data transformation rules and schema mappings","optional: reconciliation rules for conflicting data from multiple sources"],"output_types":["unified portfolio data (holdings, prices, performance)","normalized market data (prices, volatilities, correlations)","data quality metrics and reconciliation reports","streaming data feeds for real-time analysis"],"categories":["tool-use-integration","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-avanzai__cap_7","uri":"capability://planning.reasoning.backtesting.and.historical.performance.analysis.with.agent.driven.optimization","name":"backtesting and historical performance analysis with agent-driven optimization","description":"Executes agent-driven backtests that replay historical market data, apply portfolio strategies (rebalancing rules, allocation changes, risk management rules), and compute historical performance metrics. Agents iteratively refine strategy parameters based on backtest results, optimizing for objectives like Sharpe ratio, maximum drawdown, or Calmar ratio. Supports walk-forward optimization to avoid overfitting and generates performance attribution by position and time period.","intents":["Validate portfolio strategies against historical data before deploying to live portfolios","Optimize rebalancing frequency, allocation thresholds, and risk management rules based on historical performance","Understand which positions and time periods drove historical returns and risks"],"best_for":["Quantitative asset managers developing and validating trading strategies","Risk managers backtesting risk management rules and stress-test scenarios","Portfolio strategists optimizing allocation policies and rebalancing rules"],"limitations":["Backtesting results are subject to look-ahead bias, survivorship bias, and data quality issues — historical performance may not predict future results","Parameter optimization risks overfitting to historical data; walk-forward testing mitigates but doesn't eliminate this risk","Backtesting assumes perfect execution and ignores transaction costs, slippage, and market impact — real performance will be worse","Computational cost scales with historical period length and strategy complexity; multi-year backtests on large portfolios may require hours"],"requires":["Historical price data for all portfolio holdings (minimum 3-5 years, ideally 10+ years)","Strategy specification (rebalancing rules, allocation rules, risk management rules)","Transaction cost assumptions (commissions, spreads, market impact)","Computational resources for optimization (GPU acceleration recommended for large-scale optimization)"],"input_types":["historical price data and returns","strategy parameters and rules","transaction cost assumptions","optimization objectives (Sharpe ratio, maximum drawdown, Calmar ratio, etc.)"],"output_types":["backtest performance metrics (returns, volatility, Sharpe ratio, maximum drawdown)","performance attribution by position and time period","optimized strategy parameters","walk-forward optimization results and out-of-sample performance"],"categories":["planning-reasoning","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-avanzai__cap_8","uri":"capability://automation.workflow.regulatory.compliance.monitoring.and.reporting","name":"regulatory compliance monitoring and reporting","description":"Deploys agents that continuously monitor portfolio compliance with regulatory constraints (position limits, sector caps, concentration rules, leverage limits, short-sale restrictions) and generates compliance reports for regulators. Agents track regulatory changes and automatically update compliance rules, flag violations, and recommend corrective actions. Integrates with portfolio data to produce audit trails and evidence of compliance.","intents":["Ensure portfolio remains compliant with regulatory constraints without manual daily monitoring","Generate compliance reports and audit trails for regulatory examinations","Receive alerts when portfolio approaches or violates regulatory limits, enabling proactive corrective action"],"best_for":["Regulated asset managers (mutual funds, hedge funds, pension funds) subject to SEC, FINRA, or other regulatory requirements","Banks and insurance companies managing regulatory capital and risk limits","Compliance teams needing automated monitoring and reporting to reduce manual work"],"limitations":["Regulatory rules are complex and jurisdiction-specific; agents must be carefully configured for each regulatory regime","Regulatory changes require manual updates to compliance rules; no automatic tracking of regulatory updates","Compliance violations may require human judgment to resolve (e.g., whether to liquidate positions or request exemption); agents can flag but not resolve","Audit trail completeness depends on data quality and logging infrastructure; gaps in data may create compliance gaps"],"requires":["Portfolio data with real-time or daily updates","Regulatory rule specifications (position limits, sector caps, concentration rules, etc.)","Audit logging infrastructure to track compliance decisions and corrective actions","Regulatory expertise to configure rules correctly for each jurisdiction"],"input_types":["portfolio composition and market data","regulatory rule specifications","optional: regulatory change notifications or updates"],"output_types":["compliance status and violation alerts","recommended corrective actions","compliance reports and audit trails","regulatory disclosures and certifications"],"categories":["automation-workflow","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"awesome-avanzai__cap_9","uri":"capability://planning.reasoning.client.preference.learning.and.personalized.allocation.recommendations","name":"client preference learning and personalized allocation recommendations","description":"Learns client risk preferences, investment goals, and constraints from historical portfolio decisions and interactions, then generates personalized allocation recommendations aligned with learned preferences. Uses preference inference (e.g., inverse optimization) to extract implicit risk aversion and return targets from past decisions, and applies learned preferences to new allocation problems. Agents adapt recommendations over time as client preferences evolve.","intents":["Generate allocation recommendations that match client risk tolerance and investment goals without requiring explicit questionnaires","Detect changes in client preferences and adapt recommendations accordingly","Explain recommendations in terms of client-specific goals and constraints"],"best_for":["Wealth advisors managing many clients and wanting to personalize recommendations at scale","Robo-advisors learning client preferences from behavior rather than questionnaires","Asset managers wanting to tailor recommendations to individual investor preferences"],"limitations":["Preference learning requires sufficient historical data (typically 10+ decisions) to be reliable; new clients have no preference history","Learned preferences may reflect past mistakes or constraints rather than true preferences; requires human validation","Preference drift over time requires continuous re-learning; old preferences may become stale","Privacy concerns with storing and analyzing client decision history; requires careful data governance"],"requires":["Historical portfolio decisions and allocation changes for each client","Client metadata (age, income, goals, constraints) for context","Preference inference algorithm (inverse optimization, preference learning, or custom logic)","Ongoing data collection of new decisions for continuous learning"],"input_types":["historical portfolio decisions and allocations","client metadata and goals","market data and performance data","optional: explicit preference statements or questionnaire responses"],"output_types":["inferred client preferences (risk aversion, return targets, constraints)","personalized allocation recommendations","confidence scores for recommendations","preference change alerts"],"categories":["planning-reasoning","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":27,"verified":false,"data_access_risk":"high","permissions":["Real-time or near-real-time market data feed (Bloomberg, Reuters, or alternative data provider)","Portfolio holdings data in structured format (CSV, JSON, or direct broker API connection)","API credentials for Avanzai platform","Minimum portfolio size/complexity threshold unknown","Current portfolio holdings with cost basis and acquisition dates","Target allocation policy or risk profile specification","Tax rate assumptions and tax-loss harvesting rules","Liquidity constraints and position limit rules (regulatory or internal)","Real-time or daily pricing for all holdings","Real-time or near-real-time portfolio pricing (intraday for active traders, daily for passive investors)"],"failure_modes":["Agentic reasoning adds latency (typically 5-30 seconds per portfolio analysis) compared to pre-computed risk models","Accuracy depends on quality and timeliness of underlying market data feeds — stale prices degrade risk estimates","Backtesting and stress-testing capabilities unknown; may not capture regime-change risks in extreme market conditions","No explicit handling of illiquid assets or private equity positions that lack continuous pricing","Constraint satisfaction complexity grows exponentially with number of positions and constraints — may timeout on very large portfolios (1000+ holdings)","Tax-loss harvesting logic depends on accurate cost-basis tracking; errors in historical data propagate to recommendations","No explicit handling of behavioral constraints (e.g., client aversion to selling winners) — requires manual override capability","Rebalancing frequency and thresholds must be pre-configured; no adaptive logic for market regime changes","Alert fatigue risk if thresholds are not carefully tuned — false positives reduce actionability","Anomaly detection baselines require sufficient historical data (typically 60+ days) to be reliable; new portfolios may generate spurious alerts","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.05,"quality":0.3,"ecosystem":0.25,"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":"2026-06-17T09:51:02.371Z","last_scraped_at":"2026-05-03T14:00:10.321Z","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=avanzai","compare_url":"https://unfragile.ai/compare?artifact=avanzai"}},"signature":"eLUS2ZwymJaOgjpTesmCXuAvAYT6FLIXS6utOrqDPsmCL+mQ4pC6Wk8L+GAN2zrP8+VSRFSCW/wxdrB5MChTAg==","signedAt":"2026-06-20T03:02:18.608Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/avanzai","artifact":"https://unfragile.ai/avanzai","verify":"https://unfragile.ai/api/v1/verify?slug=avanzai","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"}}