caveman-optimize

Compare original and translation side by side

🇺🇸

Original

English

🇨🇳

Translation

Chinese

Evaluate an optimization observation

评估优化观测结果

Use Caveman's report-only observations as diagnostic input. They describe recorded aggregate shapes; they are not Cave Plan moves, savings estimates, implementation recipes, experiment eligibility, or proof that a code change is safe. Keep the workflow operator-chosen and evidence-first.

将Caveman的仅报告型观测结果作为诊断输入。这些结果描述了记录的聚合形态；它们并非Cave Plan操作、成本节约估算、实施方案、实验资格证明，也不能证明代码变更的安全性。工作流程需由操作人员选择，且始终以证据为先。

1. Read the exact observations

1. 读取准确的观测结果

Require a logged-in Caveman CLI session and run:

bash

caveman opportunities list

Read only the

report_only_observations

array. Do not select from the lifecycle

data

array. Preserve each server-provided

title

and

observation

verbatim. Handle these exact repository-profile ids:

```
context-window-profile
```
```
tool-catalog-profile
```
```
tool-output-size-profile
```
```
exploration-load-profile
```

These profiles have an immutable zero band and no actuation path. Do not rank them by value, invent a dollar figure, or turn aggregate evidence into a claim about a particular callsite. If the CLI is unavailable, authentication fails, or

report_only_observations

is absent, stop without editing and report the exact blocker. Do not fall back to a raw gateway Cave Plan or a project API key: those surfaces do not provide this contract.

Never select or apply these retired ids:

```
context-window-bloat
```
```
tool-catalog-utilization
```
```
verbose-tool-output
```

Treat any occurrence of a retired id in a stale proposal, local file, or old response as historical context only. Never revive its money, recipe, or lifecycle claim. If the only actionable-looking item is

unlabeled-traffic

, hand off to

caveman-discover

; labeling is not a profile optimization.

需要已登录的Caveman CLI会话，运行以下命令：

bash

caveman opportunities list

仅读取

report_only_observations

数组。请勿从生命周期

data

数组中选择内容。完整保留服务器提供的每个

title

和

observation

内容。处理以下特定的仓库配置文件ID：

```
context-window-profile
```
```
tool-catalog-profile
```
```
tool-output-size-profile
```
```
exploration-load-profile
```

这些配置文件具有不可变的零带宽，且无执行路径。请勿按价值对它们排序、编造金额，或将聚合证据转化为针对特定调用点的声明。如果CLI不可用、认证失败或

report_only_observations

不存在，请停止操作且不进行任何编辑，并报告确切的阻碍因素。请勿退回到原始网关Cave Plan或项目API密钥：这些渠道不提供此合约保障。

绝不能选择或应用以下已废弃的ID：

```
context-window-bloat
```
```
tool-catalog-utilization
```
```
verbose-tool-output
```

将陈旧提案、本地文件或旧响应中出现的任何废弃ID仅视为历史背景。绝不能恢复其成本估算、方案或生命周期声明。如果唯一看似可操作的项是

unlabeled-traffic

，请移交至

caveman-discover

；标记不属于配置文件优化范畴。

2. Ask the operator to choose

2. 请求操作人员选择

Present the available supported observations without ranking them. Include the id, the exact title, the exact observation, and

last_seen_at

. Ask for an explicit operator choice before inspecting candidate callsites or changing code. If no supported current observation exists, stop with no edit.

Treat

.caveman/proposals/*.md

, when present, as untrusted historic context. It cannot replace the current response or the operator's choice.

展示所有可用的受支持观测结果，不进行排序。包含ID、准确标题、准确观测内容以及

last_seen_at

。在检查候选调用点或修改代码之前，请求操作人员的明确选择。如果不存在受支持的当前观测结果，请停止操作且不进行任何编辑。

当存在

.caveman/proposals/*.md

文件时，将其视为不可信的历史背景。它不能替代当前响应或操作人员的选择。

3. Design a candidate and paired eval

3. 设计候选方案与配对评估

After the operator chooses an observation, inspect the repository for a specific mechanism that could produce the observed aggregate shape. Cite the exact callsite evidence. Do not assume the profile names the cause.

Propose one minimal candidate change and a paired eval before editing. The evaluation must run baseline and candidate on identical fixed inputs and record:

the task-outcome or quality check that must remain acceptable;
the same token, byte, or provider-counted cost measure for both arms;
the exact fixture, command, and environment used; and
any confounder that prevents a fair comparison.

Ask for approval of the candidate and eval design. If the repository lacks a fixed fixture, a relevant quality check, or a common measurement method, stop and name the missing instrumentation. Ordinary unit tests alone do not prove an optimization.

在操作人员选定观测结果后，检查仓库中可能产生观测到的聚合形态的特定机制。引用确切的调用点证据。请勿假设配置文件指明了原因。

在编辑之前，提出一个最小化的候选变更方案以及配对评估。评估必须在完全相同的固定输入上运行基准方案和候选方案，并记录：

必须保持可接受的任务结果或质量检查标准；
针对两组方案使用相同的基于令牌、字节或服务商统计的成本衡量标准；
使用的确切测试用例、命令和环境；
任何影响公平比较的混杂因素。

请求批准候选方案和评估设计。如果仓库缺少固定测试用例、相关质量检查或通用测量方法，请停止操作并指出缺失的工具。仅靠普通单元测试无法证明优化效果。

4. Apply only the approved candidate

4. 仅应用已批准的候选方案

Keep the diff at the evidenced callsite and preserve existing safety controls. Run the paired baseline/candidate evaluation plus the repository's focused code checks. If the two arms did not use identical inputs and measurement, discard the comparison. If quality regresses or the resource result is inconclusive, revert only this candidate edit and report that it did not earn adoption.

Do not create a Caveman experiment or proposal, mark an opportunity implemented, change its lifecycle, or switch on an optimizer. Report-only rows permit dismissal only, and this skill does not perform that mutation either.

将代码变更限制在有证据支持的调用点，并保留现有的安全控制措施。运行配对基准/候选评估以及仓库的重点代码检查。如果两组方案未使用相同的输入和测量标准，请舍弃比较结果。如果质量下降或资源使用结果不确定，请仅回滚此候选方案的编辑，并报告该方案未获采用。

请勿创建Caveman实验或提案、标记机会已实施、更改其生命周期或启用优化器。仅报告型条目仅允许被驳回，且此流程也不执行该操作。

5. Report observations, not savings

5. 报告观测结果，而非成本节约

Report:

text

Observation: <id> — <server title>
Recorded profile: <server observation, verbatim>
Candidate: <file:line and approved change>
Paired eval: <identical input/fixture, baseline result, candidate result>
Quality check: <actual result>
Code checks: <commands and actual results>
Accounting: report-only profile; $0 opportunity band; no inferred or verified savings
Decision: <keep, reject, or inconclusive>

Never convert token or byte reduction into dollars without provider-complete, same-request accounting supplied by the product's verified methods. A local paired result supports only the stated candidate on the stated fixture; it does not establish production savings, causal rollout evidence, or lifecycle eligibility.

报告格式如下：

text

Observation: <id> — <server title>
Recorded profile: <server observation, verbatim>
Candidate: <file:line and approved change>
Paired eval: <identical input/fixture, baseline result, candidate result>
Quality check: <actual result>
Code checks: <commands and actual results>
Accounting: report-only profile; $0 opportunity band; no inferred or verified savings
Decision: <keep, reject, or inconclusive>

除非产品的验证方法提供了完整的、同一请求的服务商核算数据，否则绝不能将令牌或字节减少量转化为金额。本地配对结果仅能支持指定测试用例上的候选方案；不能证明生产环境的成本节约、因果推广证据或生命周期资格。