Detect cost anomalies

Explains unusual spending through task volume, retries, model mix, prompt growth and cache behavior.

cost-anomaly-detection

Team members using this skill

AgentMeter

See how it helps

When is it useful?

Use when AI costs rise unexpectedly and a total alone cannot explain the cause.

What does it help you do?

Compares spending velocity and efficiency ratios against relevant historical baselines.
Breaks increases down by tenant, run, model and operation.
Produces diagnostic findings that can inform existing bounded execution controls.

Try asking this

Investigate yesterday's AI cost spike, separating increased demand from repeated calls, larger prompts and reduced cache reuse. Compare equal task volumes across the two days. Identify the specific runs responsible for excess cost and distinguish new traffic from repeated provider calls.

Detailed documentation

Read the skill’s original instructions, examples and checks in their authored language.

Details…Show less

Cost Anomaly Detection

Sudden cost spikes rarely look like normal growth. They usually stem from a specific pathology: an infinite retry loop, a broken prompt cache prefix, a runaway subagent tree, or an unauthenticated endpoint leak.

Workflow

Monitor structural efficiency ratios alongside raw spend:
- Track per-model, per-agent, and per-tenant health indicators over rolling windows (5m, 1h, 24h):
  - Prompt cache hit ratio: $\text{cache_read_input_tokens} / \text{total_input_tokens}$. A sudden drop from 85% to 5% after a deploy indicates a dynamic timestamp or UUID was injected near the start of the system prompt.
  - Cost per completed task: $\text{Total Run Cost} / \text{Completed Tasks}$. Spikes indicate excessive turns, tool-call loops, or high failure/retry rates.
  - Retry amplification factor: $\text{Total Provider Calls} / \text{Logical Steps}$.
  - Input-to-output token ratio: Unusually large input growth across consecutive turns within a single runId signals unbounded context accumulation or huge tool outputs being dumped into context.
Baseline comparison and velocity checks:
- Compare current window spend velocity (microsPerMinute) against:
  - Rolling historical baseline for the same tenant/workflow (seasonality-aware day-of-week / hour-of-day median and MAD/IQR).
  - Absolute velocity circuit breakers (e.g., a single task run exceeding $X in 60 seconds or N consecutive identical tool calls).
Produce actionable diagnostic payloads:
- When flagging an anomaly, never emit a vague "Cost is high" alert. Include the top contributing dimensions:
  - Top tenantId, taskId / runId, agentRole, model, and operationType responsible for the delta.
  - Decomposition of the spike: volume change (more requests), mix shift (switching from a small model to a frontier model), size inflation (larger prompts/outputs), or cache collapse (lost cache hits).
Coordinate automated containment:
- Feed runaway loop signals (maxTurnsExceeded, repeated identical tool errors, velocity breach) into the budget reservation circuit breaker to pause the offending runId or apiKeyId without taking down healthy tenants.

Anti-Patterns to Avoid

Alerting only on daily/monthly totals: Waiting 24 hours to detect a runaway loop that spends $50/hour.
Global-only thresholds: A $10/hour spike on a small workspace gets masked by aggregate platform traffic unless baselines are evaluated per tenant and per workflow.

Included files 1

SKILL.md