Prebuilt skills for Adaptive Data Collector
Adaptive Data Collector captures context‑aware diagnostic data after a performance degradation event. This targeted data collection provides relevant insights while avoiding continuous, high‑volume data collection, minimizing system impact.
Agent-based skills
The following table lists the prebuilt skills available for this agent:
| Skill name | Product name | Model | Description |
|---|---|---|---|
| Parsing a four-row hierarchy | BMC Helix Operations Management | Azure OpenAI deployment (configured in config.json, current gpt-5-mini) | Interprets metrics across a fixed four-level topology to preserve context from infrastructure to deep components. |
Detecting breaches by using percentile thresholds | BMC Helix Operations Management | Azure OpenAI deployment (configured in config.json, current gpt-5-mini) | Identifies anomalies by comparing metric behavior against learned percentile-based baselines rather than static limits. |
| Establishing a correlation between health, process, and deep monitoring | BMC Helix AIOps | Azure OpenAI deployment (configured in config.json, current gpt-5-mini) | Relates high-level health indicators with process-level and deep-dive metrics to identify meaningful dependencies. |
| Detecting the lag relation | BMC Helix Operations Management | Azure OpenAI deployment (configured in config.json, current gpt-5-mini) | Determines time-lagged relationships between metrics to understand delayed cause-and-effect behavior. |
Estimating causal strength by using causation-scoring heuristics | BMC Helix AIOps | Azure OpenAI deployment (configured in config.json, current gpt-5-mini) | Scores the likelihood and strength of causal relationships by using heuristic-based causation models. |
| Recommending an adaptive poll interval | BMC Helix Operations Management | Azure OpenAI deployment (configured in config.json, current gpt-5-mini) | Dynamically adjusts metric collection frequency based on data volatility and monitoring relevance. |
Grouping and deduplication of actionable attributes | BMC Helix Operations Management | Azure OpenAI deployment (configured in config.json, current gpt-5-mini) | Groups related signals and removes duplicates to surface only unique, action-worthy attributes. |
Performance reporting and scale extrapolation | BMC Helix Operations Management | Azure OpenAI deployment (configured in config.json, current gpt-5-mini) | Analyzes observed performance trends and projects behavior under increased scale conditions. |
Prompt-based skills
The following table lists the prebuilt prompt-based skills available for this agent:
| Skill name | Product name | Model | Description |
| Detect breached health attributes and windows | BMC Helix Operations Management | Azure OpenAI deployment (configured in config.json, current gpt-5-mini) | Identifies health attributes that exceed acceptable thresholds and determines the time windows in which breaches occur. |
| Correlate process metrics with breached health signals | BMC Helix Operations Management | Azure OpenAI deployment (configured in config.json, current gpt-5-mini) | Links process-level metrics to health attribute breaches to provide contextual insight into potential causes. |
| Trigger deep monitor selection when SQL/MSSQL context appears | BMC Helix Operations Management | Azure OpenAI deployment (configured in config.json, current gpt-5-mini) | Detects SQL or MSSQL-related signals and initiates selection of relevant deep-dive monitors. |
| Generate policy mappings for monitor enablement | BMC Helix Operations Management | Azure OpenAI deployment (configured in config.json, current gpt-5-mini) | Produces policy-to-monitor mappings to support automated or guided monitor activation. |
| Recommend polling intervals from volatility/anomaly patterns | BMC Helix Operations Management | Azure OpenAI deployment (configured in config.json, current gpt-5-mini) | Suggests optimal polling frequencies based on observed metric volatility and anomaly behavior. |
Related topics
Investigating a sudden performance degradation by using the Adaptive Data Collector