static audit · 2026-10-25T22:34:09.827Z

geekan/MetaGPT

120 source files scanned · 6 LLM call site(s) · ~7% of LLM spend looks recoverable (static heuristics)

badge   ![TokenZero audit](https://tokenzero.sh/v1/badge/audit/aud_J-dL8H06CpJnVYgbbaCROw.svg)

repo badge   [![LLM spend audited by TokenZero](https://tokenzero.sh/v1/badge/repo/geekan/MetaGPT.svg)](https://tokenzero.sh/audit/geekan/MetaGPT)

Models referenced

none detected

Findings

SeverityWhereFindingEst.
mediummetagpt/prompts/di/engineer2.py:620 static prompt template(s), ~13,382 tokens, and no caching directive anywhere in the repo
Most call sites use providers with automatic prefix caching, which only pays off when these templates are sent as a byte-identical prefix: static template first, nothing volatile (dates, ids, user data) interpolated into it.
Fix: Keep templates first and static; TokenZero adds a stable `prompt_cache_key` so related requests land on the same cache.		12%

Fix all of it with one line

OPENAI_BASE_URL=https://tokenzero.sh/v1
ANTHROPIC_BASE_URL=https://tokenzero.sh

Keys pass through untouched. Cache formatting and exact caching are on by default; cascade routing stays opt-in. How it works