perf(review): replace N verification agents with single batch verification

Previously, each finding got its own independent verification agent
(N findings = N LLM calls). Now a single verification agent receives
all findings at once and verifies them in one pass.

Token cost: 6+N variable calls → 7 fixed calls (5 review + 1 verify + 1 reverse audit)
Quality: minimal impact — batch verification has fuller cross-finding context

Co-Authored-By: Claude Opus 4.6 (1M context) <noreply@anthropic.com>
This commit is contained in:
wenshao 2026-04-07 03:19:23 +08:00
parent 08a797cf76
commit e65e5bd353
2 changed files with 6 additions and 8 deletions

View file

@ -49,7 +49,7 @@ Step 5: Clean up (remove worktree and temp files)
| Agent 4: Undirected Audit | Business logic, boundary interactions, hidden coupling |
| Agent 5: Build & Test | Runs build and test commands, reports failures |
All agents run in parallel. Each finding is independently verified by a separate verification agent to reduce false positives. After verification, a **reverse audit agent** reviews the diff with knowledge of all confirmed findings to catch issues that every other agent missed.
All agents run in parallel. All findings are then verified in a **single batch verification pass** (one agent reviews all findings at once, keeping LLM calls fixed regardless of finding count). After verification, a **reverse audit agent** reviews the diff with knowledge of all confirmed findings to catch issues that every other agent missed.
## Deterministic Analysis