cost-booster-route

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Cost Booster Route

Cost Booster 路由

Wraps

mcp__claude-flow__hooks_route

and reports how many tasks the 3-tier router classified as Agent Booster (Tier 1) eligible. Tier 1 bypasses run as WASM transforms — no LLM call, structurally $0 cost.

封装

mcp__claude-flow__hooks_route

，并报告3层路由器将多少任务归类为符合Agent Booster（Tier 1）条件的任务。Tier 1旁路以WASM转换的形式运行——无需调用LLM，从结构上实现0美元成本。

When to use

使用场景

Before a batch of similar tasks, or when

cost-report

shows Sonnet/Opus spend on descriptions that look like simple transforms (

var-to-const

add-types

add-error-handling

async-await

add-logging

remove-console

在处理一批相似任务之前，或者当

cost-report

显示Sonnet/Opus在看似简单转换的描述上产生支出时（例如

var-to-const

、

add-types

、

add-error-handling

、

async-await

、

add-logging

、

remove-console

）。

Steps

步骤

Collect tasks — single arg or recent entries from
```
cost-tracking
```
via
```
memory_search
```
. Cap batch at 50.
Route each — call
```
hooks_route
```
with the description; capture the full response string.
Partition — group by whether the response contains the literal
```
[AGENT_BOOSTER_AVAILABLE]
```
(Tier 1) vs. not (Tier 2/3). Extract
```
[TASK_MODEL_RECOMMENDATION] Use model="X"
```
when present.
Compute spend — Tier 1 partition: $0. Tier 2/3 partition: per-task upper-bound from REFERENCE.md pricing × recommended model.

Report

=== Booster bypass report ===
Tasks analyzed:        50
Tier 1 (booster):      18 (36%)  — $0.00
Tier 2/3 (LLM):        32 (64%)  — $X.XX (upper-bound)
Booster intents:       var-to-const (8), add-types (5), remove-console (5)

Persist —

memory_store --namespace cost-tracking --key "booster-route-$(date +%Y%m%d-%H%M%S)" --value '{"tier1": N, "tier2_or_3": M, ...}'

cost-report

picks up the tier signal.

收集任务 —— 单个参数或通过
```
memory_search
```
从
```
cost-tracking
```
获取近期条目。批量上限为50个。
逐个路由 —— 使用任务描述调用
```
hooks_route
```
；捕获完整响应字符串。
分区 —— 根据响应是否包含字面量
```
[AGENT_BOOSTER_AVAILABLE]
```
（Tier 1）进行分组，不包含则归为Tier 2/3。若存在
```
[TASK_MODEL_RECOMMENDATION] Use model="X"
```
则提取该内容。
计算支出 —— Tier 1分区：0美元。Tier 2/3分区：基于REFERENCE.md中的定价上限×推荐模型，计算每个任务的支出上限。

生成报告

=== Booster旁路报告 ===
分析任务数:        50
Tier 1（Booster）:      18 (36%)  — $0.00
Tier 2/3（LLM）:        32 (64%)  — $X.XX (上限)
Booster意图:       var-to-const (8), add-types (5), remove-console (5)

持久化存储 —— 执行

memory_store --namespace cost-tracking --key "booster-route-$(date +%Y%m%d-%H%M%S)" --value '{"tier1": N, "tier2_or_3": M, ...}'

，以便

cost-report

能获取到分层信号。

Caveats — claimed upstream, not yet verified

注意事项——上游声称，尚未验证

```
[AGENT_BOOSTER_AVAILABLE]
```
fires only when the upstream router populates
```
routeResult.agentBoosterIntent.type
```
(
```
v3/@claude-flow/cli/src/mcp-tools/hooks-tools.ts:1228
```
). The published CLI's semantic-VectorDb path does not always trigger the classifier — treat the partition as a lower bound on Tier 1 eligibility.
CLAUDE.md root claims
```
<1ms
```
latency and
```
352× faster
```
than LLM.
```
<1ms
```
and
```
$0
```
are structural;
```
352×
```
is claimed upstream, not yet verified here. Report what the router actually returns.
See
```
docs/benchmarks/0002-baseline.md
```
for the full upstream-claims-vs-measured table.

```
[AGENT_BOOSTER_AVAILABLE]
```
仅在上游路由器填充
```
routeResult.agentBoosterIntent.type
```
时触发（路径：
```
v3/@claude-flow/cli/src/mcp-tools/hooks-tools.ts:1228
```
）。已发布的CLI的语义向量数据库路径并不总是触发分类器——请将分区结果视为Tier 1合格性的下限。
CLAUDE.md根文档声称延迟
```
<1ms
```
，比LLM快352倍。
```
<1ms
```
和
```
0美元
```
是结构特性；352倍是上游声称的数据，尚未在此处验证。请报告路由器实际返回的结果。
完整的上游声称与实测数据对比表请查看
```
docs/benchmarks/0002-baseline.md
```
。

Cross-references

交叉引用

ADR-0002 Decision #1 · ruflo-intelligence ADR-0001 §"Neutral" (closes the routing-outcomes loop via

cost-optimize

step 8) · CLAUDE.md root §"3-Tier Model Routing (ADR-026)".

ADR-0002 决策#1 · ruflo-intelligence ADR-0001 §"Neutral"（通过

cost-optimize

步骤8关闭路由结果循环）· CLAUDE.md根文档 §"3-Tier Model Routing (ADR-026)"。