Resources

Agent Status / Field Notes

Agent Reliability Metrics Reference

Complete guide to agent reliability metrics: definitions, calculations, and usage.


01Section

Metric Categories

CategoryPurpose
AvailabilityIs the agent reachable?
CorrectnessIs the agent giving right answers?
PerformanceIs the agent fast enough?
OperationalHow well are we responding to issues?
BusinessWhat's the user/business impact?
02Section

Availability Metrics

Uptime

Definition: Percentage of time the agent endpoint returns a valid response.

snippet
Uptime = (Successful Checks / Total Checks) × 100

Successful check: HTTP 2xx response received within timeout.

Use CaseTarget
Internal tool99%
Production API99.5%
Critical service99.9%

Availability

Definition: Percentage of time the agent is both reachable AND functional.

snippet
Availability = (Checks Passed / Total Checks) × 100

Where "passed" means:
- HTTP response received (2xx or handled 4xx)
- Response within timeout
- Evaluation prompts passed (if configured)

Distinction from Uptime: Uptime asks "Does it respond?" - Availability asks "Does it work?"

Effective Uptime

Definition: Availability accounting for semantic correctness.

snippet
Effective Uptime = Uptime × (Eval Pass Rate / 100)

An agent returning 200 OK with wrong answers has high uptime but low effective uptime.


03Section

Correctness Metrics

Eval Pass Rate

Definition: Percentage of evaluation checks returning correct answers.

snippet
Eval Pass Rate = (Correct Gold Responses / Total Evaluation Prompts) × 100

What makes a response "correct":

  • Health tier: Expected answer appears somewhere in response
  • Contract tier: Response matches expected format exactly
TierTarget
Basic90%
Standard95%
Premium98%

Health Tier Pass Rate

Pass rate for health-tier evaluation prompts only. These test basic reasoning with loose matching - the expected answer can appear anywhere in the response.

Contract Tier Pass Rate

Pass rate for contract-tier evaluation prompts only. These test format/schema validation with exact match or structured comparison.


04Section

Performance Metrics

TTFB (Time to First Byte)

Definition: Time from request initiation until first byte of response received.

For streaming agents, TTFB is when user sees "typing..." - the primary user experience metric for streaming.

PercentileTarget
P50<2s
P95<5s
P99<10s

Total Latency

Definition: Time from request initiation until complete response received.

snippet
Total Latency = DNS + TCP + TLS + Server Processing + Transfer

Latency Percentiles

P50 (Median): Half of requests faster, half slower. Represents "typical" experience.

P95: 95% of requests faster. Common SLA target.

P99: 99% of requests faster. Reveals tail latency.

Why percentiles over averages:

snippet
Example requests (ms): 100, 100, 100, 100, 5000

Average: 1080ms (misleading)
P50: 100ms (typical experience)
P99: 5000ms (worst case visible)

05Section

Operational Metrics

MTTD (Mean Time to Detect)

snippet
MTTD = Sum(Detection Time - Start Time) / Incident Count
LevelTarget
Basic<30 min
Good<15 min
Excellent<5 min

MTTR (Mean Time to Resolution)

snippet
MTTR = Sum(Resolution Time - Detection Time) / Incident Count
SeverityTarget
Critical<1 hour
High<4 hours
Medium<24 hours

Change Failure Rate

snippet
CFR = (Deployments Causing Incidents / Total Deployments) × 100
LevelTarget
Concerning>10%
Acceptable5-10%
Good<5%
Elite<1%

06Section

Verdicts

Verdict Definitions

UP: Availability ≥95%, Eval Pass Rate ≥80%, Latency within SLA, Majority of tests successful.

DEGRADED: Between UP and DOWN thresholds. Partial functionality, some tests failing, elevated latency.

DOWN: Availability <20%. Majority of tests failing, complete or near-complete outage.

UNKNOWN: Insufficient data, all tests failed, or initial state before first check.

Verdict Calculation

snippet
if probe_pass_rate >= 95% AND gold_pass_rate >= 80% AND latency_ok:
    verdict = UP
elif probe_pass_rate < 20%:
    verdict = DOWN
else:
    verdict = DEGRADED

System Verdict

TARGET_EVALUATED: Test successfully reached agent. Result reflects actual agent state. Counts toward availability.

VALIDATOR_ERROR: Test failed due to monitoring infrastructure. Not agent's fault. Excluded from availability.


07Section

Error Budget

snippet
Error Budget = 100% - SLA Target

For 99.9% SLA:
Error Budget = 0.1%
Per month (43,200 minutes): 43.2 minutes allowed

Error Budget Policy

Budget healthy (>50% remaining): Normal operations, deployments allowed.

Budget low (<25% remaining): Increased caution, limit risky changes.

Budget exhausted: Stop non-essential changes, reliability focus only.


08Section

Metric Anti-Patterns

Average-Only Latency

Averages hide tail latency. Always use percentiles (P50, P95, P99).

Uptime Without Correctness

Misses "up but broken" state. Include gold prompt validation.

Single-Location Metrics

Misses regional failures. Track per-region and global separately.

Point-in-Time Only

Misses trends and intermittent issues. Track over time, analyze trends.


09Section

Dashboard Recommendations

Real-Time Dashboard

Current verdict per agent, last check time, rolling 1-hour availability, current TTFB P50/P95, active alerts. Refresh every 1-5 minutes.

SLA Dashboard

Period availability vs target, error budget remaining, eval pass rate, latency percentiles vs SLA, incident count. Month-to-date with historical comparison.

Trend Dashboard

Availability over time (30/60/90 days), latency trends, incident frequency, change failure rate, MTTR trend. Purpose: spot degradation before incidents.

Independent monitoring

See your agent the way the world sees it.

Outside-in validations from real residential nodes, evaluation prompts that catch silent-200 failures.