{"passport":{"unfragile":{"@version":"1.0","version":"2026-05","artifact":{"id":"athina-ai","slug":"athina-ai","name":"Athina AI","type":"dataset","url":"https://athina.ai","page_url":"https://unfragile.ai/athina-ai","categories":["observability","model-training","testing-quality"],"tags":[],"pricing":{"model":"free","free":true,"starting_price":null},"status":"active","verified":false},"capabilities":[{"id":"athina-ai__cap_0","uri":"capability://data.processing.analysis.preset.evaluation.metrics.execution","name":"preset-evaluation-metrics-execution","description":"Executes 50+ pre-built evaluation metrics (Ragas-based and custom) against LLM outputs without requiring metric implementation. Metrics include RagasAnswerCorrectness, RagasContextPrecision, RagasContextRelevancy, RagasContextRecall, RagasFaithfulness, ResponseFaithfulness, Groundedness, ContextSufficiency, DoesResponseAnswerQuery, ContextContainsEnoughInformation, and Faithfulness. Integrates with external LLM providers (OpenAI confirmed) to compute metric scores in parallel batches with configurable concurrency (max_parallel_evals parameter).","intents":["I want to evaluate RAG pipeline quality without implementing custom metrics","I need to measure hallucination, context relevance, and answer correctness across a dataset","I want to run 50+ evaluation metrics in parallel against my LLM outputs"],"best_for":["data scientists evaluating RAG systems","teams building LLM applications without ML expertise","QA teams validating response quality at scale"],"limitations":["Metric implementations are opaque — cannot customize scoring logic within preset metrics","Requires external LLM API access (OpenAI confirmed, others unknown) for metric computation","No offline evaluation — all metrics require live API calls, adding latency and cost","Preset metrics are fixed to Ragas framework definitions — cannot modify metric thresholds or weights"],"requires":["Python 3.7+","API key for OpenAI or supported LLM provider","Athina Python SDK (athina.evals, athina.runner modules)","Structured dataset with query, context, and response fields"],"input_types":["structured dataset (query, context, response tuples)","LLM outputs (text)","evaluation configuration (metric selection, concurrency settings)"],"output_types":["metric scores (numeric)","evaluation results (structured JSON)","comparison reports (side-by-side metric analysis)"],"categories":["data-processing-analysis","testing-quality"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__cap_1","uri":"capability://data.processing.analysis.custom.evaluation.metric.definition","name":"custom-evaluation-metric-definition","description":"Allows teams to define custom evaluation metrics beyond the 50+ presets by implementing metric logic that integrates with the EvalRunner orchestration system. Custom metrics are stored in Athina's platform and versioned alongside datasets and prompts. Implementation approach unknown but likely supports Python function definitions or declarative metric schemas that hook into the parallel evaluation pipeline.","intents":["I need to evaluate LLM outputs against domain-specific criteria not covered by preset metrics","I want to define a custom metric for my specific use case and reuse it across evaluation runs","I need to version and track changes to my custom metric definitions"],"best_for":["teams with specialized evaluation requirements","domain experts defining industry-specific quality criteria","organizations building proprietary evaluation frameworks"],"limitations":["Custom metric implementation details and API surface unknown — insufficient documentation","No visibility into how custom metrics integrate with parallel execution — potential performance unknowns","Custom metrics are locked into Athina platform — no export or reuse outside the platform","Unclear whether custom metrics can call external APIs or are limited to local computation"],"requires":["Python 3.7+","Athina Python SDK","Understanding of metric interface/contract (undocumented)"],"input_types":["metric definition (Python function or schema)","evaluation data (query, context, response)"],"output_types":["metric score (numeric)","evaluation result (structured)"],"categories":["data-processing-analysis","testing-quality"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__cap_10","uri":"capability://tool.use.integration.external.llm.provider.integration.and.key.management","name":"external-llm-provider-integration-and-key-management","description":"Integrates with external LLM providers (OpenAI confirmed, others unknown) to execute evaluations and run AI workflows. Manages API keys securely via AthinaApiKey.set_key() and OpenAiApiKey.set_key() methods. Abstracts provider-specific API differences, allowing teams to swap models without changing evaluation code. Handles API rate limiting, retries, and error handling transparently.","intents":["I want to use OpenAI models for evaluation without managing API calls directly","I need to switch between different LLM providers without rewriting evaluation code","I want to securely store and manage API keys for multiple providers"],"best_for":["teams using multiple LLM providers","engineers building provider-agnostic evaluation pipelines","organizations with security requirements around API key management"],"limitations":["Supported LLM providers beyond OpenAI unknown","API key storage and encryption approach not documented — unclear if keys are stored encrypted or in plaintext","No mention of key rotation, expiration, or audit logging","Rate limiting and retry logic are opaque — no configuration options mentioned","Unclear whether custom LLM providers (e.g., self-hosted models) are supported"],"requires":["API key for supported LLM provider (OpenAI confirmed)","Python SDK (athina.keys module)","Network access to provider APIs"],"input_types":["API key (string)","provider selection (e.g., 'openai')"],"output_types":["authenticated provider connection (used internally by Athina)"],"categories":["tool-use-integration","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__cap_11","uri":"capability://data.processing.analysis.evaluation.dataset.loading.and.transformation","name":"evaluation-dataset-loading-and-transformation","description":"Provides loaders (athina.loaders.Loader) to import evaluation datasets from various sources (CSV, JSON, API, pre-built datasets like yc_query_mini) and transform them into Athina's internal format. Loaders handle schema mapping, data validation, and format conversion. Pre-built datasets are available for quick prototyping. Supports programmatic dataset construction via Python tuples or objects.","intents":["I want to load my evaluation dataset from a CSV file into Athina","I need to quickly prototype with a pre-built dataset without preparing my own","I want to transform my dataset schema to match Athina's expected format"],"best_for":["data engineers preparing evaluation datasets","teams prototyping with pre-built datasets","organizations migrating datasets from external sources"],"limitations":["Supported data sources and formats unknown — only CSV, JSON, and pre-built datasets mentioned","Schema mapping and validation rules not documented","No mention of data quality checks or validation error reporting","Unclear whether loaders support streaming large datasets or require full in-memory loading","No mention of data transformation capabilities (filtering, aggregation, sampling)"],"requires":["Python 3.7+","Athina Python SDK (athina.loaders module)","Dataset in supported format (CSV, JSON, or pre-built)"],"input_types":["dataset file (CSV, JSON)","dataset reference (pre-built dataset name)","programmatic dataset (Python tuples or objects)"],"output_types":["loaded dataset (in Athina's internal format)","validation errors (if schema mismatch)"],"categories":["data-processing-analysis","tool-use-integration"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__cap_12","uri":"capability://memory.knowledge.evaluation.run.history.and.artifact.tracking","name":"evaluation-run-history-and-artifact-tracking","description":"Maintains a complete history of evaluation runs, including metadata (timestamp, user, configuration), input datasets, metrics, and results. Each run is linked to specific prompt versions, model selections, and retriever configurations, creating an audit trail. Teams can retrieve past runs, compare results, and reproduce evaluations. Likely uses a database to store run metadata and results with queryable indexes.","intents":["I want to see the history of all evaluation runs and who triggered them","I need to reproduce an evaluation from 2 weeks ago with the same configuration","I want to track which prompt/model/retriever combination was used in each run"],"best_for":["teams with governance and compliance requirements","organizations tracking evaluation history for audits","engineers debugging evaluation issues by reviewing past runs"],"limitations":["Run history retention policy unknown — unclear how long runs are stored","No mention of run archival, export, or deletion capabilities","Unclear whether run history can be queried programmatically or only via UI","No mention of run comparison or diff capabilities","Storage limits and quotas unknown"],"requires":["Athina account","Evaluation runs to have been executed and stored"],"input_types":["run ID or date range (for retrieval)","filter criteria (user, configuration, metric)"],"output_types":["run metadata (timestamp, user, configuration)","run results (metrics, sample-level scores)","run artifacts (datasets, prompts, models used)"],"categories":["memory-knowledge","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__cap_13","uri":"capability://data.processing.analysis.metric.score.aggregation.and.statistical.analysis","name":"metric-score-aggregation-and-statistical-analysis","description":"Aggregates metric scores across evaluation samples and computes statistical summaries (mean, standard deviation, percentiles, min/max). Supports filtering and grouping by dimensions (e.g., by sample type, query length, retriever). Likely uses NumPy or similar for efficient computation. Enables teams to understand metric distributions and identify outliers.","intents":["I want to see the mean and standard deviation of my evaluation metrics","I need to identify samples where metrics are outliers (very high or very low)","I want to group metric results by sample type to see if quality varies"],"best_for":["data scientists analyzing metric distributions","teams identifying quality issues in specific sample subsets","organizations tracking metric trends and variability"],"limitations":["Supported aggregation functions and grouping dimensions unknown","No mention of percentile calculation or outlier detection methods","Unclear whether statistical tests (t-tests, ANOVA) are supported","No mention of visualization of distributions (histograms, box plots)","Grouping dimensions and custom aggregations not documented"],"requires":["Athina account","Completed evaluation run with metric scores"],"input_types":["evaluation results (metric scores per sample)","aggregation configuration (grouping dimensions, functions)"],"output_types":["aggregated statistics (mean, std, percentiles, min/max)","grouped results (statistics per group)","outlier identification (samples with extreme scores)"],"categories":["data-processing-analysis","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__cap_2","uri":"capability://data.processing.analysis.dataset.curation.and.versioning","name":"dataset-curation-and-versioning","description":"Manages evaluation datasets with versioning, annotation, and SQL-based querying capabilities. Datasets are stored in Athina's platform with version history, enabling teams to track changes and regenerate datasets by modifying model, prompt, or retriever configurations. Includes pre-built datasets (e.g., yc_query_mini) and loaders for importing external data. Supports side-by-side dataset comparison with SQL query interface for data scientists.","intents":["I want to curate and version evaluation datasets without managing files","I need to compare two versions of a dataset side-by-side to understand what changed","I want to regenerate a dataset by changing the model or retriever and see the impact","I need to query and filter datasets using SQL for exploratory analysis"],"best_for":["data scientists managing evaluation datasets","teams collaborating on dataset curation","organizations tracking dataset lineage and versioning"],"limitations":["Dataset format and schema are proprietary to Athina — no export format mentioned, creating vendor lock-in","SQL query interface is limited to data scientists — no visual query builder for non-technical users mentioned","Dataset regeneration requires re-running inference through external LLM providers, incurring API costs","No mention of dataset size limits, storage quotas, or retention policies","Annotation workflow details unknown — unclear if human annotation is built-in or requires external tools"],"requires":["Athina account with dataset management permissions","Python SDK (athina.loaders, athina.datasets modules) for programmatic access","Structured data in query/context/response format or custom schema"],"input_types":["structured data (CSV, JSON, or programmatic tuples)","dataset configuration (model, prompt, retriever settings)","SQL queries for filtering/analysis"],"output_types":["versioned dataset (stored in Athina)","comparison reports (side-by-side diffs)","query results (filtered dataset subsets)"],"categories":["data-processing-analysis","memory-knowledge"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__cap_3","uri":"capability://automation.workflow.batch.evaluation.execution.with.parallelization","name":"batch-evaluation-execution-with-parallelization","description":"Orchestrates batch evaluation runs across multiple metrics and dataset samples using parallel execution with configurable concurrency (max_parallel_evals parameter). EvalRunner.run_suite() method accepts a list of evaluation metrics, a dataset, and concurrency settings, then distributes evaluation work across worker threads/processes. Results are aggregated and returned as structured evaluation reports. Handles API rate limiting and error handling for external LLM provider calls.","intents":["I want to evaluate 1000+ samples against 10+ metrics in reasonable time","I need to control parallelism to avoid hitting API rate limits","I want to run evaluation suites programmatically without manual orchestration"],"best_for":["teams evaluating large datasets (1000+ samples)","engineers building evaluation pipelines","organizations running nightly evaluation jobs"],"limitations":["Parallelization adds complexity to debugging — failures in parallel workers may be difficult to trace","max_parallel_evals parameter requires manual tuning based on API rate limits and quota","No mention of distributed execution across multiple machines — parallelization appears to be thread/process-based on single machine","Evaluation latency depends on external LLM API response times — no SLA or timeout guarantees mentioned","No built-in retry logic or fault tolerance mentioned — transient API failures may cause entire batch to fail"],"requires":["Python 3.7+","Athina Python SDK (athina.runner.run module)","API keys for external LLM providers","Dataset loaded into memory or streamed from Athina"],"input_types":["list of evaluation metrics (Eval objects)","dataset (structured tuples or Athina dataset reference)","concurrency configuration (max_parallel_evals integer)"],"output_types":["evaluation results (structured JSON with metric scores per sample)","aggregated statistics (mean, std, percentiles per metric)"],"categories":["automation-workflow","data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__cap_4","uri":"capability://safety.moderation.real.time.application.monitoring.and.quality.detection","name":"real-time-application-monitoring-and-quality-detection","description":"Monitors LLM-powered applications in production to detect quality degradation, hallucinations, and context relevance issues in real-time. Integrates with running applications to capture LLM inputs/outputs and compute evaluation metrics continuously. Detects anomalies such as response quality drops, increased hallucination rates, or context mismatches. Implementation details unknown but likely uses streaming evaluation and statistical anomaly detection.","intents":["I want to detect when my LLM application's response quality degrades in production","I need to identify hallucinations as they occur in live user interactions","I want alerts when context relevance drops below acceptable thresholds"],"best_for":["teams running LLM applications in production","organizations with SLAs on response quality","product teams monitoring user-facing AI features"],"limitations":["Real-time monitoring implementation details completely unknown — unclear how metrics are computed without adding latency to user requests","No mention of alert configuration, thresholds, or notification channels","Unclear whether monitoring requires code instrumentation or works via API interception","No SLA or latency guarantees for anomaly detection mentioned","Cost implications unknown — real-time evaluation likely incurs per-request charges to external LLM providers"],"requires":["Athina account with monitoring enabled","Integration with running LLM application (method unknown)","API keys for external LLM providers (for metric computation)"],"input_types":["LLM application inputs (queries, prompts)","LLM application outputs (responses)","monitoring configuration (metrics, thresholds, alert rules)"],"output_types":["quality metrics (real-time scores)","anomaly alerts (threshold violations)","monitoring dashboards (time-series visualizations)"],"categories":["safety-moderation","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__cap_5","uri":"capability://memory.knowledge.multi.model.prompt.management.and.comparison","name":"multi-model-prompt-management-and-comparison","description":"Manages and versions prompts across multiple LLM providers (OpenAI confirmed, others unknown) with side-by-side comparison and evaluation capabilities. Teams can test the same prompt against different models (e.g., GPT-4 vs GPT-3.5) and compare results. Prompts are versioned in Athina's platform and linked to evaluation runs, enabling teams to track which prompt version produced which results. Supports prompt templates with variable substitution.","intents":["I want to test the same prompt against GPT-4 and GPT-3.5 and compare results","I need to version my prompts and track which version was used in each evaluation run","I want to compare prompt variations side-by-side to see which performs better"],"best_for":["prompt engineers optimizing LLM outputs","teams comparing model performance","organizations managing prompt libraries across multiple models"],"limitations":["Supported LLM providers beyond OpenAI are unknown","Prompt template syntax and variable substitution rules not documented","No mention of prompt versioning branching or merging — appears to be linear history only","Unclear whether prompts can be exported or are locked into Athina platform","No mention of prompt optimization or automated prompt tuning"],"requires":["Athina account","API keys for LLM providers being compared","Python SDK or web UI for prompt management"],"input_types":["prompt text (with optional template variables)","model selection (e.g., gpt-4, gpt-3.5-turbo)","evaluation dataset"],"output_types":["versioned prompts (stored in Athina)","comparison reports (side-by-side results across models)","evaluation metrics per prompt/model combination"],"categories":["memory-knowledge","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__cap_6","uri":"capability://data.processing.analysis.retriever.configuration.and.evaluation","name":"retriever-configuration-and-evaluation","description":"Allows teams to configure and evaluate different retrieval strategies (e.g., different vector databases, chunking strategies, embedding models) and measure their impact on RAG pipeline quality. Datasets can be regenerated by changing retriever configuration, enabling A/B testing of retrieval approaches. Evaluation metrics like RagasContextPrecision and RagasContextRelevancy measure retrieval quality. Implementation details unknown but likely supports pluggable retriever interfaces.","intents":["I want to compare different vector databases or chunking strategies for my RAG pipeline","I need to measure how retriever changes impact context relevance and answer correctness","I want to A/B test different embedding models without manual dataset regeneration"],"best_for":["teams optimizing RAG pipelines","engineers evaluating vector database options","organizations tuning retrieval strategies"],"limitations":["Supported retriever types and vector databases unknown","Retriever configuration interface and API not documented","No mention of retriever performance metrics (latency, throughput) — only quality metrics","Unclear whether custom retrievers can be plugged in or if only pre-built integrations are supported","Dataset regeneration with different retrievers requires re-running inference, incurring API costs"],"requires":["Athina account","Configured retriever (vector database, embedding model, chunking strategy)","API keys for external LLM providers (for evaluation metrics)"],"input_types":["retriever configuration (vector database, embedding model, chunk size, etc.)","evaluation dataset"],"output_types":["context relevance scores (RagasContextRelevancy, RagasContextPrecision)","context recall scores (RagasContextRecall)","comparison reports (retriever A vs retriever B)"],"categories":["data-processing-analysis","search-retrieval"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__cap_7","uri":"capability://planning.reasoning.no.code.ai.flow.prototyping","name":"no-code-ai-flow-prototyping","description":"Enables non-technical users (product managers, business analysts) to prototype multi-step AI workflows without code. Provides a visual interface for chaining prompts, models, and retrievers together. Workflows can be tested against datasets and evaluated using preset metrics. Implementation details unknown but likely uses a DAG-based flow editor with drag-and-drop components.","intents":["I want to prototype an AI workflow without writing code","I need to test a multi-step prompt chain against my evaluation dataset","I want to see how different model choices affect my workflow output"],"best_for":["non-technical product managers","business analysts prototyping AI features","teams with mixed technical/non-technical stakeholders"],"limitations":["No-code flow editor capabilities and limitations completely unknown","Unclear what types of operations can be chained (prompts, retrievers, custom logic, etc.)","No mention of conditional logic, loops, or error handling in flows","Unclear whether flows can be exported or are locked into Athina platform","No mention of performance optimization or flow execution latency"],"requires":["Athina account with web UI access","No coding required, but understanding of AI workflow concepts helpful"],"input_types":["visual flow definition (drag-and-drop components)","model/prompt/retriever selections"],"output_types":["workflow definition (stored in Athina)","workflow execution results (outputs per step)","evaluation metrics (applied to final output)"],"categories":["planning-reasoning","automation-workflow"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__cap_8","uri":"capability://data.processing.analysis.human.annotation.and.labeling.workflow","name":"human-annotation-and-labeling-workflow","description":"Supports human annotation of evaluation datasets alongside automated metrics, enabling teams to create ground truth labels for model evaluation. Annotators can review LLM outputs and provide feedback (e.g., correctness, relevance, hallucination presence). Annotations are stored in Athina and can be used to validate automated metric accuracy. Implementation details unknown but likely includes annotation UI, reviewer management, and inter-rater agreement tracking.","intents":["I want to have humans label my evaluation dataset for ground truth","I need to validate whether my automated metrics agree with human judgment","I want to track annotation progress and manage multiple annotators"],"best_for":["QA teams validating LLM outputs","organizations building labeled datasets","teams measuring metric accuracy against human judgment"],"limitations":["Annotation workflow UI and capabilities completely unknown","No mention of inter-rater agreement metrics or conflict resolution","Unclear whether annotations can be exported or are locked into Athina","No mention of annotation guidelines, templates, or quality control mechanisms","Cost and pricing for human annotation services unknown"],"requires":["Athina account with annotation feature enabled","Evaluation dataset loaded into Athina","Human annotators with Athina account access"],"input_types":["LLM outputs (query, context, response)","annotation guidelines (optional)","annotation schema (label types, options)"],"output_types":["human labels (stored in Athina)","annotation metadata (annotator, timestamp, confidence)","inter-rater agreement metrics (if multiple annotators)"],"categories":["data-processing-analysis","safety-moderation"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__cap_9","uri":"capability://data.processing.analysis.evaluation.result.comparison.and.reporting","name":"evaluation-result-comparison-and-reporting","description":"Generates side-by-side comparison reports of evaluation runs, enabling teams to understand how changes (prompt, model, retriever) impact metric scores. Reports show metric deltas, statistical significance (if applicable), and sample-level breakdowns. Supports filtering and sorting by metric, sample, or other dimensions. Likely uses statistical aggregation and visualization to surface insights.","intents":["I want to compare evaluation results from two different prompts side-by-side","I need to see which samples improved and which regressed when I changed the model","I want to understand the statistical significance of metric changes"],"best_for":["data scientists analyzing evaluation results","teams making model/prompt decisions","organizations tracking metric trends over time"],"limitations":["Statistical significance testing approach unknown — unclear if p-values, confidence intervals, or other methods are used","Report customization and export options unknown","No mention of time-series analysis or trend detection","Unclear whether reports can be shared with non-technical stakeholders","No mention of automated insights or recommendations based on comparisons"],"requires":["Athina account","At least two evaluation runs to compare","Access to evaluation results (stored in Athina)"],"input_types":["evaluation run IDs or date ranges","filter/sort criteria (metric, sample, dimension)"],"output_types":["comparison report (side-by-side metrics)","metric deltas (absolute and percentage changes)","sample-level breakdowns (which samples improved/regressed)","statistical analysis (significance, confidence intervals, etc.)"],"categories":["data-processing-analysis","planning-reasoning"],"confidence":0.5,"matches":0,"success_rate":0},{"id":"athina-ai__headline","uri":"capability://data.processing.analysis.llm.evaluation.and.monitoring.platform","name":"llm evaluation and monitoring platform","description":"Athina AI is an advanced evaluation and monitoring platform specifically designed for LLM-powered applications, providing real-time monitoring, custom evaluation metrics, and dataset curation to ensure high-quality AI outputs.","intents":["best LLM evaluation platform","monitoring tool for LLM applications","evaluation metrics for AI models","real-time monitoring for language models","dataset curation for LLMs"],"best_for":["data scientists","QA teams","product managers"],"limitations":["dependent on external APIs","evaluation scope may vary"],"requires":["API integration","dataset input"],"input_types":["datasets","API responses"],"output_types":["evaluation reports","monitoring alerts"],"categories":["data-processing-analysis"],"confidence":0.5,"matches":0,"success_rate":0}],"trust":{"score":58,"verified":false,"data_access_risk":"high","permissions":["Python 3.7+","API key for OpenAI or supported LLM provider","Athina Python SDK (athina.evals, athina.runner modules)","Structured dataset with query, context, and response fields","Athina Python SDK","Understanding of metric interface/contract (undocumented)","API key for supported LLM provider (OpenAI confirmed)","Python SDK (athina.keys module)","Network access to provider APIs","Athina Python SDK (athina.loaders module)"],"failure_modes":["Metric implementations are opaque — cannot customize scoring logic within preset metrics","Requires external LLM API access (OpenAI confirmed, others unknown) for metric computation","No offline evaluation — all metrics require live API calls, adding latency and cost","Preset metrics are fixed to Ragas framework definitions — cannot modify metric thresholds or weights","Custom metric implementation details and API surface unknown — insufficient documentation","No visibility into how custom metrics integrate with parallel execution — potential performance unknowns","Custom metrics are locked into Athina platform — no export or reuse outside the platform","Unclear whether custom metrics can call external APIs or are limited to local computation","Supported LLM providers beyond OpenAI unknown","API key storage and encryption approach not documented — unclear if keys are stored encrypted or in plaintext","builder identity is not verified yet","no observed match outcomes yet"],"rank_breakdown":{"adoption":0.7,"quality":0.9,"ecosystem":0.35,"match_graph":0.25,"freshness":0.75,"weights":{"adoption":0.3,"quality":0.25,"ecosystem":0.1,"match_graph":0.3,"freshness":0.05}},"observed_outcomes":{"matches":0,"success_rate":0,"avg_confidence":0,"top_intents":[],"last_matched_at":null},"maintenance":{"status":"active","updated_at":"2026-05-24T12:16:19.836Z","last_scraped_at":null,"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=athina-ai","compare_url":"https://unfragile.ai/compare?artifact=athina-ai"}},"signature":"P6hOXccetkPJj5hnBnwB2wjtM+yWE2mBiCP8R5ANCQ5VBm8/NAZ1G41omhSsMvZ0JMn3QzNPhCdhHJkjUHMvAw==","signedAt":"2026-06-19T21:42:05.779Z","signedBy":"unfragile.ai","version":1},"_links":{"self":"https://unfragile.ai/api/v1/passport/athina-ai","artifact":"https://unfragile.ai/athina-ai","verify":"https://unfragile.ai/api/v1/verify?slug=athina-ai","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"}}