Skip to main content
Tero analyzes your data and surfaces rules. Each rule is a specific statement about what’s wrong: this field duplicates that one, this log is a health check probe, this event fires a million times but represents one problem. Rules are organized by category. Each category is a type of mistake, something concrete that shouldn’t be in your data.

​
The trust ladder

Tero organizes issues by risk level. Start at the top and work your way down.

Zero risk

You’ll agree immediately: redundant attributes, leftover debug logs, malformed data

Low risk

Straightforward but worth a glance: health checks, bot traffic, unintended tool metadata

Medium risk

Requires discussion with the team: excessive payloads, debug mode left on, sampling decisions
Pick a category, look at examples from your data, approve if you agree. You review categories, not individual rules. Your ruleset grows over time.

​
Categories

​
Logs

Some categories affect attributes (fields within a log). Others affect entire events. Attribute-level — the log stays, specific fields get removed:

Redundant attributes

Two fields with the same value

Unintended tool metadata

Fields added by agents you never search

Excessive payloads

Response bodies and large nested objects
Event-level — the entire log gets dropped or sampled:

Malformed data

Binary blobs, corrupted output

Leftover debug logs

Debug statements that shipped by accident

Debug mode left on

Verbose logging never disabled

Logs in hot path

Log statements in middleware or tight loops

Health checks

Readiness probes, liveness checks

Bot traffic

Crawlers, scrapers, automated scanners

Excessive repetition

High volume where cost outweighs value

PII leakage

Sensitive data that ended up in logs

​
Metrics

High cardinality tags

Tags with unbounded values that explode storage

​
Enforcement

Once you approve a rule, you choose how it gets enforced: configure your provider directly, push to the edge, open PRs, create tickets. If a problem resurfaces — an engineer re-enables debug logging, a new service emits the same waste pattern — Tero catches it. Problems you fix stay fixed. See enforcement options →