Files
memind/docs/goose-v149-real-llm-gate.md
T
john 2fa9523587
Memind CI / Test, build, and release guards (push) Has been cancelled
feat(goose-v149): add offline phase3 gate without real LLM calls
Block unattended DashScope smokes by default and provide a zero-token
merge verification entry via run-goosed-v149-phase3-offline.mjs.

Co-authored-by: Cursor <cursoragent@cursor.com>
2026-09-09 22:10:19 +08:00

2.1 KiB
Raw Blame History

Goose v1.49 真实 LLM 测试闸门(2026-09-09 收口)

背景

本机 :18049 v1.49 回归(check-goosed-v149-all / run-goosed-v149-phase3 等)在一天内可创建 1000+ Goose session、消耗 1 亿+ DashScope token,主要来自:

  • Portal smoke / resume(完整 Agent 多轮 + 工具)
  • Page E2E(苏州一日游等生页场景)
  • Memory loop / chat

默认禁止再跑这类会调用真实模型的聚合 smoke;只允许显式 opt-in。

默认行为(收口后)

动作 无 opt-in GOOSE_V149_ALLOW_REAL_LLM=1
bash scripts/run-goosed-v149-local.sh 拒绝启动 允许
node scripts/check-goosed-v149-all.mjs 立即退出 2 允许
node scripts/run-goosed-v149-phase2.mjs 立即退出 2 允许
node scripts/run-goosed-v149-phase3.mjs 立即退出 2 允许
node scripts/run-goosed-v149-missing-evidence.mjs 立即退出 2 允许
单测 / message sanitize / memory manifest 对照 仍允许 仍允许

Portal / page / resume smoke 在未 opt-in 时还会被强制 *_SKIP=1(即使单独跑子脚本)。

显式批准后再跑(一次性)

# 1. 仅在需要时启动 v1.49 goosed
GOOSE_V149_ALLOW_REAL_LLM=1 bash scripts/run-goosed-v149-local.sh

# 2. 另开终端,跑指定检查(勿循环 / 勿 CI 默认开启)
GOOSE_V149_ALLOW_REAL_LLM=1 node scripts/check-goosed-v149-reply-smoke.mjs

禁止Agent/Cursor 在未获用户明确批准时设置 GOOSE_V149_ALLOW_REAL_LLM=1 或循环重跑 phase2/phase3。

推荐替代

  • 零 LLM 合并前验证(默认推荐):
node scripts/run-goosed-v149-phase3-offline.mjs
  • 单项:node --test scripts/goose-v149-canary.test.mjscompare-goose-v149-message-sanitize.mjscompare-goose-v149-memory-manifest.mjs
  • 必须验证 Portal 集成时:单次、有人值守、先确认 DashScope 配额,且 GOOSE_V149_ALLOW_REAL_LLM=1

回滚 Canary

若主 Portal 曾切到 :18049

bash scripts/switch-goose-v149-canary.sh rollback
# 重启 pnpm dev