{ "format": "daily-project-prototype-v1", "project": "ai-observability", "title": "Production AI Observability Monitor", "domain": "ai-observability", "mode": "classifier", "labels": [ "latency-regression", "token-spike", "tool-failure", "quality-regression" ], "prototypes": { "latency-regression": { "request": 3, "latency": 6, "rose": 3, "above": 3, "the": 9, "service": 3, "objective": 3, "end": 6, "to": 3, "exceeded": 3, "approved": 3, "percentile": 3, "threshold": 3, "in": 2, "an": 4, "operations": 2, "review": 2, "for": 2, "evaluation": 2, "case": 2, "agent": 3, "stayed": 3, "within": 3, "quality": 6, "limits": 3, "but": 3, "became": 3, "slower": 3, "performance": 3, "changed": 3, "without": 3, "a": 3, "matching": 3, "improvement": 3 }, "token-spike": { "in": 1, "an": 2, "operations": 1, "review": 1, "prompt": 2, "tokens": 2, "doubled": 2, "after": 2, "a": 2, "template": 2, "change": 2, "token": 2, "consumption": 2, "increased": 2, "beyond": 2, "the": 2, "cost": 2, "and": 2, "context": 2, "baseline": 2, "for": 1, "evaluation": 1, "case": 1 }, "tool-failure": { "the": 4, "retrieval": 2, "tool": 4, "returned": 2, "a": 4, "timeout": 2, "exception": 2, "required": 2, "external": 2, "failed": 4, "during": 2, "execution": 2, "for": 2, "an": 3, "evaluation": 2, "case": 2, "in": 1, "operations": 1, "review": 1, "search": 2, "calls": 2, "with": 2, "repeated": 2, "connection": 2, "errors": 4, "dependency": 2, "prevented": 2, "workflow": 2, "from": 2, "completing": 2 }, "quality-regression": { "grounded": 2, "answer": 2, "score": 2, "dropped": 2, "after": 2, "deployment": 2, "evaluation": 2, "quality": 2, "regressed": 2, "relative": 2, "to": 2, "the": 2, "release": 2, "baseline": 2, "in": 1, "an": 1, "operations": 1, "review": 1 } }, "idf": { "end": 2.252763, "to": 1.847298, "latency": 2.252763, "exceeded": 2.252763, "the": 1.336472, "approved": 2.252763, "percentile": 2.252763, "threshold": 2.252763, "token": 2.252763, "consumption": 2.252763, "increased": 2.252763, "beyond": 2.252763, "cost": 2.252763, "and": 2.252763, "context": 2.252763, "baseline": 1.847298, "a": 1.847298, "required": 2.252763, "external": 2.252763, "tool": 2.252763, "failed": 2.252763, "during": 2.252763, "execution": 2.252763, "evaluation": 2.252763, "quality": 1.847298, "regressed": 2.252763, "relative": 2.252763, "release": 2.252763, "performance": 2.252763, "changed": 2.252763, "without": 2.252763, "matching": 2.252763, "improvement": 2.252763, "dependency": 2.252763, "errors": 2.252763, "prevented": 2.252763, "workflow": 2.252763, "from": 2.252763, "completing": 2.252763 }, "documents": [ { "id": "trace-01", "label": "latency-regression", "text": "End-to-end latency exceeded the approved percentile threshold.", "metadata": { "synthetic": true, "domain": "ai-observability" } }, { "id": "trace-02", "label": "token-spike", "text": "Token consumption increased beyond the cost and context baseline.", "metadata": { "synthetic": true, "domain": "ai-observability" } }, { "id": "trace-03", "label": "tool-failure", "text": "A required external tool failed during execution.", "metadata": { "synthetic": true, "domain": "ai-observability" } }, { "id": "trace-04", "label": "quality-regression", "text": "Evaluation quality regressed relative to the release baseline.", "metadata": { "synthetic": true, "domain": "ai-observability" } }, { "id": "trace-05", "label": "latency-regression", "text": "Performance changed without a matching quality improvement.", "metadata": { "synthetic": true, "domain": "ai-observability" } }, { "id": "trace-06", "label": "tool-failure", "text": "Dependency errors prevented the workflow from completing.", "metadata": { "synthetic": true, "domain": "ai-observability" } } ], "graph_edges": [], "confidence_threshold": 0.18, "trained_on_synthetic_data": true }