Skill UI

This skill runs comprehensive benchmarks against structural and adversarial corpora. It measures the performance and cost efficiency of a booster model compared to leading LLM baselines (Gemini, Anthropic). Results are persisted in verifiable JSON reports, crucial for auditing claims and verifying booster win rates before publishing.

A high-performance utility for executing simple, deterministic code transformations (e.g., removing console logs, converting var to const). This engine leverages a WASM core to perform structural refactoring sub-millisecond, without calling any expensive LLM APIs. Ideal for CI/CD pipelines, automated bulk refactoring, and cost-sensitive development workflows where determinism and speed are paramount.

This tool wraps the hooks routing mechanism to classify tasks into tiers based on Agent Booster eligibility. It calculates potential cost savings by identifying Tier 1 bypass tasks (which cost $0) versus standard LLM calls. It is essential for optimizing large batches of simple, predictable operations and reporting accurate resource utilization.

This skill monitors project spending against a configured budget limit. It aggregates total costs from session tracking records, calculates the utilization percentage, and emits a multi-tiered alert (INFO, WARNING, CRITICAL, HARD_STOP). It is essential for cost governance, ensuring that resource spawning (like swarms) is halted or optimized when spending approaches critical thresholds.

Monitors the acceleration trend of production spending (burn rate) over defined time windows. It compares the latest spending bucket against the historical average, providing an early warning system. This is crucial for detecting sudden, unexpected cost spikes, such as a 'hot loop,' independent of overall budget limits.

This skill wraps the `getCompactContext` functionality to efficiently retrieve a compacted ReasoningBank context, specifically optimized for cost-analysis queries. It generates a detailed report summarizing critical metrics such as tokens saved, memories retrieved, and agentic-flow availability. It is essential for monitoring and optimizing resource usage in complex AI workflows, providing deep visibility into the efficiency and cost of the context retrieval mechanism.

This skill provides a granular, session-by-session view of total accumulated costs. It lists metrics such as start time, message count, top model, and total cost for every recorded conversation. This is ideal for identifying which specific sessions consumed the most resources, which is crucial for project retrospectives and evaluating the sustainability of current session patterns.

Performs multi-baseline counterfactual cost analysis by comparing actual session spending against hypothetical costs generated by fixed, non-optimal models (e.g., always-haiku, always-sonnet). This tool quantifies the financial savings realized by the smart routing mechanism, answering the critical question: 'Is the routing earning its keep?' It is ideal for quarterly cost reviews, setting CI performance gates, and validating routing configurations.

This tool detects cost regressions by comparing two cost-summary JSON snapshots (baseline vs. current). It is essential for PR-gate workflows, checking for unexpected spending increases using three orthogonal signals: total percentage change, absolute dollar increase, and disproportionate growth in specific token classes (e.g., cache writes).

This skill exports cost-tracking telemetry data into formats suitable for external observability platforms like Prometheus, Grafana, and Datadog. It allows cost data, previously confined to the AgentDB namespace, to be visualized and monitored in real-time dashboards via Prometheus textfile collectors or structured JSON webhooks.

This skill consumes federation_spend events to aggregate per-peer spending across rolling time windows (1h, 24h, 7d). It implements a crucial circuit breaker mechanism by monitoring if any peer's spending exceeds a predefined threshold, ensuring the stability and resource management of the federation. Used to establish spending baselines and verify budget constraints.

This composite CI gate runs four critical cost monitoring checks—budget, burn, anomaly, and projection—in parallel. It surfaces a single, combined health status by taking the maximum exit code. This tool is an operational entry point for integrating comprehensive resource spending alerts directly into CI/CD pipelines, ensuring that resource usage remains within defined boundaries and predicts potential budget overruns.