cost-compact-context

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Cost Compact Context

成本压缩上下文

Wraps

getTokenOptimizer().getCompactContext()

from

@claude-flow/integration

for cost-analysis queries. The bridge dynamically imports

agentic-flow

with graceful fallback: when the package isn't installed,

tokensSaved

and the skill exits cleanly. No MCP tool wraps

getTokenOptimizer

today (ADR-0002 §"Riskiest assumption"); we shell a Node one-liner instead.

封装来自

@claude-flow/integration

的

getTokenOptimizer().getCompactContext()

，用于成本分析查询。桥接器会动态导入

agentic-flow

并提供优雅降级方案：当该包未安装时，

tokensSaved

设为

，技能会平稳退出。目前没有MCP工具封装

getTokenOptimizer

（ADR-0002 §"风险最高的假设"）；因此我们改用Node单行命令实现。

Steps

步骤

Take the query — the single argument.

Invoke — run from anywhere under

v3/

@claude-flow/integration

resolves:

bash

( cd v3 && node ../plugins/ruflo-cost-tracker/scripts/compact.mjs "<QUERY>" )

The script imports

@claude-flow/integration/token-optimizer

(canonical export — not

dist/token-optimizer.js

, which would double the

.js

extension via Node's

./*

exports rule), calls

getCompactContext(query)

, and prints a markdown summary plus a JSON line via

COMPACT_QUIET=1

Report — markdown table with: memories retrieved, tokens saved (bridge-reported), agentic-flow availability, cache hit rate. The script also emits a "bridge-reported, not measured against a no-RAG baseline" disclaimer. On bridge-unavailable: prints "agentic-flow not installed — bridge returns inert results." and exits cleanly.

接收查询 —— 唯一的参数。
调用执行 —— 在
```
v3/
```
目录下任意位置运行，确保
```
@claude-flow/integration
```
能正常解析：
bash
```
( cd v3 && node ../plugins/ruflo-cost-tracker/scripts/compact.mjs "<QUERY>" )
```
该脚本导入
```
@claude-flow/integration/token-optimizer
```
（标准导出 —— 不要使用
```
dist/token-optimizer.js
```
，否则会因Node的
```
./*
```
导出规则导致
```
.js
```
扩展名重复），调用
```
getCompactContext(query)
```
，并通过
```
COMPACT_QUIET=1
```
输出markdown摘要和一行JSON数据。
结果上报 —— 生成包含以下内容的markdown表格：检索到的记忆数量、桥接器返回的token节省量、agentic-flow可用性、缓存命中率。脚本还会附带一条“数据来自桥接器，未与无RAG基线对比”的声明。当桥接器不可用时：打印“agentic-flow未安装 —— 桥接器返回无效结果。”并平稳退出。

Caveats — claimed upstream, not yet verified

注意事项 —— 上游声称，尚未验证

CLAUDE.md root claims

ReasoningBank retrieval: -32%

tokens. The bridge's

tokensSaved

query_tokens − compact_prompt_tokens

(token-optimizer.ts:141–143) — a heuristic, not a baseline-measured saving. token-optimizer.ts:9–10 itself says: "No fabricated metrics are reported — all stats reflect real measurements". This skill carries that disclaimer forward.

Booster-specific availability is not exposed as a getter — observable only through

optimizedEdit()

returning

method: 'agent-booster'

. The canonical Tier 1 signal is

[AGENT_BOOSTER_AVAILABLE]

(see

cost-booster-route

CLAUDE.md根文档声称

ReasoningBank检索：token减少32%

。桥接器的

tokensSaved

计算方式为

query_tokens − compact_prompt_tokens

（token-optimizer.ts:141–143）—— 这是一种启发式算法，并非基于基线测量的真实节省量。token-optimizer.ts:9–10本身说明：“不报告虚构指标 —— 所有统计数据均反映真实测量结果”。本技能会沿用该声明。

Booster的具体可用性未通过getter暴露 —— 仅能通过

optimizedEdit()

method: 'agent-booster'

观察到。标准的Tier 1信号是

[AGENT_BOOSTER_AVAILABLE]

（参见

cost-booster-route

）。

Fallback

降级方案

agentic-flow

not installed →

getCompactContext

returns

{tokensSaved: 0, memories: []}

(line 116–124),

optimizedEdit

returns

{method: 'traditional'}

getOptimalConfig

falls back to anti-drift defaults. Skill exits cleanly with the "not available" message.

若

agentic-flow

未安装 →

getCompactContext

{tokensSaved: 0, memories: []}

（第116–124行），

optimizedEdit

{method: 'traditional'}

，

getOptimalConfig

回退到防漂移默认值。技能会打印“不可用”提示并平稳退出。

Cross-references

交叉引用

ADR-0002 Decision #2 + §"Riskiest assumption" ·

token-optimizer.ts:308

(singleton export) ·

docs/benchmarks/0002-baseline.md

(verification findings).

ADR-0002 决策#2 + §"风险最高的假设" ·

token-optimizer.ts:308

（单例导出） ·

docs/benchmarks/0002-baseline.md

（验证结果）。