op1 营销 seo Exper

安全社区

作者仓库星标 0
作者仓库 skills-registry

解读按原文结构重写，命令、链接、术语均保留；右侧可核对作者原始 SKILL.md

原文结构

Objective：Run the Stage1 SEO pipeline in shadow mode and produce auditable artifacts under research/stage1marketingseo/.
Single-cycle run：From workspace root: Check: research/stage1marketingseo/run.latest.json
Guardrails：Keep mode as shadow for Stage1 capability validation. Treat this as experiment generation/scoring/queueing only (not auto publish). If run status is blockedinputincomplete, stop and report missing requirements from run.latest.json.inputsync.completeness.

适用与边界

当前原文没有单列适用场景。
1/2/3 validation run：Use the bundled helper: This executes: Round1 baseline

原文中的明确线索

要点：「This executes: 1.」、「--- > Source: Daihaolin201/OperatorOne — distributed by TomeVault.」
文件与命令：research/stage1marketingseo/、research/stage1marketingseo/run.latest.json、research/stage1marketingseo/experiments.queue.latest.json、research/stage1marketingseo/scoreboard.latest.json、research/stage1marketingseo/decisionlog.latest.md、research/stage1marketingseo/rounds/<timestamp>/rounds123.report.md、research/stage1marketingseo/rounds/<timestamp>/rounds123.report.json、R1 == R3

流狐整理：以上内容来自当前 SKILL.md 的章节与原词；未补写作者没有声明的工具、兼容性或能力。

Stage1 SEO Experiment Runner (Shadow)

Objective

Run the Stage1 SEO pipeline in shadow mode and produce auditable artifacts under research/stage1_marketing_seo/.

Single-cycle run

From workspace root:

./scripts/run_marketing_seo_stage1.sh --mode shadow --force

Check:

research/stage1_marketing_seo/run.latest.json
research/stage1_marketing_seo/experiments.queue.latest.json
research/stage1_marketing_seo/scoreboard.latest.json
research/stage1_marketing_seo/decision_log.latest.md

1/2/3 validation run

Use the bundled helper:

python3 {baseDir}/scripts/run_rounds_123.py

This executes:

Round1 baseline
Round2 controlled handoff perturbation
Round3 baseline restore + rerun

Then it writes snapshot artifacts and validation checks to:

research/stage1_marketing_seo/rounds/<timestamp>/rounds_1_2_3.report.md
research/stage1_marketing_seo/rounds/<timestamp>/rounds_1_2_3.report.json

Expected checks:

R1 == R3 for queue/content reproducibility
R2 != R1 for input-sensitivity

Guardrails

Keep mode as shadow for Stage1 capability validation.
Treat this as experiment generation/scoring/queueing only (not auto publish).
If run status is blocked_input_incomplete, stop and report missing requirements from run.latest.json.input_sync.completeness.

References

references/stage1-output-contract.md
references/analysis-checklist.md

Source: Daihaolin201/OperatorOne — distributed by TomeVault.

流狐档案作者与许可取自来源；运行、权限和网络为流狐检测或估算

流狐分类: 安全
作者声明 Agent: 未找到明确声明；不据此推断已兼容或已测试
静态检查: 88 / 100 · 启发式扫描，不代表运行安全
作者 / 版本 / 许可: @tomevault-io · 未声明 license
流狐 Token 估算: 低消耗
流狐接入估算: 需简单配置
是否需要外部 API Key: 未发现要求
检测到的系统要求: 未声明
底层运行要求: 未声明
检测到的文件与系统行为: 只读
Shell 执行
检测到的网络行为: 仅限本地
安装命令数: 无（仅作为资料）

档案由构建时根据 SKILL.md 与安装命令自动衍生，可能与作者实际意图存在差异。

需要注意： 未限定 allowed-tools，默认拥有全部工具权限。

输出预览 op1-marketing-seo-experiment-stage1.preview

作者没有在当前 SKILL.md 中定义固定输出样例。

SKILL.md SKILL.md

# Stage1 SEO Experiment Runner (Shadow)

## Objective
Run the Stage1 SEO pipeline in shadow mode and produce auditable artifacts under `research/stage1_marketing_seo/`.

## Single-cycle run
From workspace root:

```bash
./scripts/run_marketing_seo_stage1.sh --mode shadow --force
```

Check:
- `research/stage1_marketing_seo/run.latest.json`
- `research/stage1_marketing_seo/experiments.queue.latest.json`
- `research/stage1_marketing_seo/scoreboard.latest.json`
- `research/stage1_marketing_seo/decision_log.latest.md`

## 1/2/3 validation run
Use the bundled helper:

```bash
python3 {baseDir}/scripts/run_rounds_123.py
```

This executes:
1. Round1 baseline
2. Round2 controlled handoff perturbation
3. Round3 baseline restore + rerun

Then it writes snapshot artifacts and validation checks to:
- `research/stage1_marketing_seo/rounds/<timestamp>/rounds_1_2_3.report.md`
- `research/stage1_marketing_seo/rounds/<timestamp>/rounds_1_2_3.report.json`

Expected checks:
- `R1 == R3` for queue/content reproducibility
- `R2 != R1` for input-sensitivity

## Guardrails
- Keep mode as `shadow` for Stage1 capability validation.
- Treat this as experiment generation/scoring/queueing only (not auto publish).
- If run status is `blocked_input_incomplete`, stop and report missing requirements from `run.latest.json.input_sync.completeness`.

## References
- `references/stage1-output-contract.md`
- `references/analysis-checklist.md`

---
> Source: [Daihaolin201/OperatorOne](https://github.com/Daihaolin201/OperatorOne) — distributed by [TomeVault](https://tomevault.io).
<!-- tomevault:4.0:skill_md:2026-05-22 -->

证据边界与执行链路

作者原文负责流程事实；流狐只索引当前章节、要点、文件与命令。

章节    -> Objective → Single-cycle run → 1/2/3 validation run → Guardrails → References
要点    -> This executes: 1. · --- > Source: [Daihaolin201/OperatorOne](https://github.com/Daihaolin201/OperatorOne) — distributed by [TomeVault](https://tomevault.io).
文件/命令 -> research/stage1marketingseo/ · research/stage1marketingseo/run.latest.json · research/stage1marketingseo/experiments.queue.latest.json · research/stage1marketingseo/scoreboard.latest.json · research/stage1marketingseo/decisionlog.latest.md · research/stage1marketingseo/rounds/<timestamp>/rounds123.report.md · research/stage1marketingseo/rounds/<timestamp>/rounds123.report.json · R1 == R3
内容 SHA-256 -> 0394b249ac2a