AI Performance

Pass-1 is the default diff review; pass-2 is an optional agentic re-review after pass-1 when enabled.

Final-pass input (code + security)

Final-pass output (code + security)

Pass-1 input (non-agentic)

Pass-1 output (non-agentic)

Pass-2 tokens (agentic)

Pass-1 share (of pass-1 + pass-2)

—

Avg core IO tokens / PR

—

Avg duration

—

Total duration (hours)

—

Cache hit rate

N/A

Est. cost

—

Re-review rate

—

Avg runs / PR

—

Code-review agentic

Activation rate

—

Avg planner iterations

—

Avg context retrieved

—

Planner token overhead

Avg planner duration

—

Security-review agentic

Activation rate

—

Avg planner iterations

—

Avg context retrieved

—

Planner token overhead

Avg planner duration

—

Rerun cost and cache behavior

Avg cost to non-blocking: — across 0 resolved PR lifecycles.

Cache hit % is cache-read tokens divided by cache read plus cache-creation tokens, split by whether the review row is the first run on a PR or a later re-review run.

Cache hit rate by run cohort: first run versus re-runs
Cohort	Cache hit rate
First runsRun index 1 on each pull request	—
Re-runsRun index 2 and above on the same PR	—

AI performance

Pass-1 is the default diff review; pass-2 is an optional agentic re-review after pass-1 when enabled.

Final-pass input (code + security)

Final-pass output (code + security)

Pass-1 input (non-agentic)

Pass-1 output (non-agentic)

Pass-2 tokens (agentic)

Pass-1 share (of pass-1 + pass-2)

—

Avg core IO tokens / PR

—

Avg duration

—

Total duration (hours)

—

Cache hit rate

N/A

Est. cost

—

Re-review rate

—

Avg runs / PR

—

Code-review agentic

Activation rate

—

Avg planner iterations

—

Avg context retrieved

—

Planner token overhead

Avg planner duration

—

Security-review agentic

Activation rate

—

Avg planner iterations

—

Avg context retrieved

—

Planner token overhead

Avg planner duration

—

Rerun cost and cache behavior

Avg cost to non-blocking: — across 0 resolved PR lifecycles.

Cache hit % is cache-read tokens divided by cache read plus cache-creation tokens, split by whether the review row is the first run on a PR or a later re-review run.

Cache hit rate by run cohort: first run versus re-runs
Cohort	Cache hit rate
First runsRun index 1 on each pull request	—
Re-runsRun index 2 and above on the same PR	—