fresh-eyes-loop SKILL.md skillA

fresh-eyes-loop from kongfangxun/sofagent is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add kongfangxun/sofagent/fresh-eyes-loop@v1.7.0

Pins v1.7.0 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
kongfangxun/sofagent
Path
FORGE/SKILL/fresh-eyes-loop/SKILL.md
Versions recorded
8
Latest version
v1.7.0, committed Oct 21, 2026
First version recorded
Oct 1, 2026
Size
5,613 bytes
Repository stars
54
License
MIT

发布后独立质量循环——单盲四角色流水线(A 审 12 视角 → B 修 → C 验 → D 复核),每轮新 session 保证零上下文,连续 2 轮无 P0/P1 即停。.

Collected from public GitHub. Every version is saved exactly as it was and never changes. Install it at an exact version with the mdr tool, and every file is checked before it reaches your agent. The main file is checked against the SHA-256 fingerprint recorded here, and the others against the fingerprints GitHub keeps for its source commit. The automatic checks below grade the latest version, and the same checks always give the same file the same grade.

What the file says

# fresh-eyes-loop · 质量循环定义

> **一个循环 = 一轮又一轮的"独立审查 → 修复 → 验证",直到干净为止。**
>
> 这不是检查清单,是一套**让独立性可被重复执行**的机制。每一轮都用全新 session 跑(零上下文),所以"作者自己看不出问题"这个人类弱点被结构性消解。

## 这是什么

一套可复用的质量循环定义。它描述:谁来做(单盲四角色:A 审 / B 修 / C 验 / D 复核)、每一轮怎么走(审查 → 修复 → 验证 → 复核)、什么时候停(连续 2 轮无 P0/P1)、产物放哪(`runs/YYYY/MM/DD/run-NN/`)。

- **A** = 审查者(单盲):独立跑 12 视角审查;发现的质量由下游 C 验收 / D 复核把关(legacy 双盲 B 并行审查走 `FORGE_ENABLE_B_CHECK=1` 逃生门)。
- **B/C/D** = 工程师执行修复(现行 B 侧为复核模式——独立复核 A 的 P0/P1,可推翻可补充)/ 验收者逐条实测验收(不采信修复自报)/ 复核者对 P0/P1 裁决 CONFIRM / DOWNGRADE / REOPEN。
- **编排者(非 agent)**:在角色间中转、维护 `runs/` 文件、判定停止条件。编排者 = **一次 harness session 本身**(WorkBuddy 有人值守 / DSH 无头 / Codex headless),注入 `loop.md`「执行形态」节的主任务协议即跑——不存在第二种编排形态(driver 多轮编排循环已随整合归一删除,经验沉淀于 `FORGE/lessons/driver.md`)。

## 怎么用

1. 读 `loop.md` 拿到完整 SOP(角色 / 轮次协议 / 产物 schema / 停止条件)。
2. 12 视角的定义见 `playbook/fresh-eyes-review.md`(A 按它跑)。playbook 共 **22 视角六层**:1-12 常规发版(driver 循环标准配置)、13-14 文档治理(手动)、15-16 全文档通读(手动/草稿工具)、**17-19 动态面**(跨组件契约/构建产物/执行证据——需跨包追踪或 build/实跑取证,worker 工具面实测仅 `sf_read` / `sf_write` / `run_bash` 三件,跨包构建取证超其超时预算,发版审查建议手动追加)、20-21 深度专项(季度全仓体检)、22 发现面(门面改动时)。
  **loop 与工具的视角边界以 playbook 分层表为准——playbook 演进(如新增视角/调整分层)时,本文件与下游工具同步对齐**。
…

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add kongfangxun/sofagent/fresh-eyes-loop@sha256:1ef894ec5f58504d

The line at the top of this page pins the label v1.7.0, which is what most people want. This one pins the sha256 fingerprint of the contents instead, so it survives a label being moved. Either way the exact version is written to mdr.lock, and mdr install downloads and checks those contents again, so it installs them exactly or stops with an error.

Badge

mdr badge

[![mdr](https://markdownregistry.com/badge/art_epknlrj6oqw5vztd.svg)](https://markdownregistry.com/a/art_epknlrj6oqw5vztd)

2 badge image fetches in 30 days. What the badge shows.

Versions

The registry has recorded 8 versions of this file, between Oct 1, 2026 and Oct 21, 2026, about one every 3 days.

versioncommittedcommitsizeaudit
v1.7.0 latest2026-10-21 0488670 5,613 BA view · diff
v1.6.02026-10-15 80f9ac2 5,613 BA view · diff
v1.5.92026-10-12 47119a7 5,613 BA view · diff
v1.5.82026-10-09 c333e50 5,613 BA view · diff
v1.5.72026-10-08 ae90dce 5,613 BA view · diff
v1.5.62026-10-04 cdc5e82 5,613 BA view · diff
v1.5.52026-10-02 bd2f7f6 5,613 BA view · diff
v1.5.52026-10-01 8937842 5,606 BA view

Audit of the latest version

A  17 of 17 checks passed. Fixed automatic checks with no AI involved, so the same checks give the same answer every run.
  • pass: Frontmatter block present
  • pass: Frontmatter declares a name
  • pass: Frontmatter declares a description
  • pass: Size between 200 bytes and 200 KB (5613 bytes)
  • pass: No zero-width or bidi control characters
  • pass: No instruction hidden inside an HTML comment
  • pass: No link to an exfiltration or paste host
  • pass: No credential-shaped string
  • pass: No instruction to send local credentials anywhere
  • pass: No text hidden with inline styles
  • pass: No prompt-injection phrasing
  • pass: No curl or wget piped into a shell
  • pass: No recursive delete of root, home or parent
  • pass: No instruction to read or print local credentials
  • pass: No base64 blob over 200 characters
  • pass: No link to a raw IP address
  • pass: No script tag

Source

GitHub

kongfangxun/sofagent · 54 stars · license MIT · pushed 2026-10-21 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_epknlrj6oqw5vztd
GET https://markdownregistry.com/api/v1/resolve?ref=kongfangxun/sofagent/fresh-eyes-loop
GET https://markdownregistry.com/api/v1/blob/1ef894ec5f58504d508d9081b1e9d1fb1353b50dae35776c454fdbb48d7000e6

Your AI assistant does the legwork. You hear about the deals worth your word. Hand yours the standing instructions at modelranch.com and it joins the network that reads files like this one.

Guides: the SKILL.md format and its required fields · what to check before you install a skill · how to pin a skill to an exact version · where to find agent skills · keeping a team on the same skills · which agent file does what.

More from kongfangxun/sofagent

sofagent Cursor rule
kongfangxun/sofagent · .cursor/rules/sofagent.mdc · sofagent 约束层加载(薄挂载——规范资产唯一来源是 SKILL/,本文件不含规则内容)
git:20261008.d73ee15 · audit A · 54 stars
AGENTS.md AGENTS.md file
kongfangxun/sofagent · AGENTS.md
git:20261008.d73ee15 · audit A · 54 stars
enterprise-fde-识别AI节点 SKILL.md skill
kongfangxun/sofagent · FDE/templates/skills/skill-template/SKILL.md · 当 FDE 需要对工作流节点做 AI 分类判定时用这个 Skill—— 不是"它是干什么的",是"什么时候用"。 写错 description = Skill 永远不会被触发。
v1.2.5 · audit A · 54 stars
release-gate-loop SKILL.md skill
kongfangxun/sofagent · FORGE/SKILL/release-gate-loop/SKILL.md · 发版前自动验证闸门——V 验证 + F 修复循环(verdict FAIL → F 改代码 → 跑 audit → V 重验),最大 3 轮直到 PASS。纯只读验证 + 最小修复。
v1.7.0 · audit A · 54 stars
AGENTS.md@SKILL AGENTS.md file
kongfangxun/sofagent · SKILL/AGENTS.md
git:20261008.02c81de · audit A · 54 stars
sofagent SKILL.md skill
kongfangxun/sofagent · SKILL/SKILL.md · FDE Skill——帮 FDE(前线部署工程师)更好完成企业 AI 落地的方法论 Skill。约束 Agent 行为、审计每次变更、沉淀经验。 底层实现叫约束层——一个层五种能力:注入·审计·回溯·沉淀·进化。FORGE…
v1.7.0 · audit A · 54 stars
sofagent-audit SKILL.md skill
kongfangxun/sofagent · SKILL/agents/audit/SKILL.md · 系统级合规审计——巡检 Workflow、验证铁律覆盖、检查知识库健康度。不审查代码逻辑,审查的是部署层面的合规性。
v1.7.0 · audit A · 54 stars
软件工程师 SKILL.md skill
kongfangxun/sofagent · SKILL/agents/engineer/SKILL.md · 专注于最小可行差异的工程专家——只修复被要求的内容,拒绝范围蔓延,宁可写三行相似代码也不做过早抽象。这种纪律性能防止 bug 修复 PR 变成重构雪崩。
v1.7.0 · audit A · 54 stars
sofagent-fde SKILL.md skill
kongfangxun/sofagent · SKILL/agents/fde/SKILL.md · 前线部署与知识工程专家。梳理企业工作流、识别 AI 节点、构建 ontology 本体数据、交付离场。 部署完成后转为持续优化模式(sustain),自动读 audit 报告趋势生成优化报告。…
v1.7.0 · audit A · 54 stars
代码审查员 SKILL.md skill
kongfangxun/sofagent · SKILL/agents/reviewer/SKILL.md · 专业代码审查专家,提供建设性、可操作的反馈,聚焦正确性、可维护性、安全性和性能,而非代码风格偏好。
v1.7.0 · audit A · 54 stars
cordis-plugin-sofagent-audit SKILL.md skill
kongfangxun/sofagent · engine/dsh-plugins/cordis-plugin-sofagent-audit/SKILL.md · 变更机器审阅 + 验收硬门禁——28 规则 + git diff 硬证据 + Turn 停止验收判定(验收不过不放行,开关独立可关)(seam: tools/result + tools/pre-execute +…
v1.7.0 · audit A · 54 stars
cordis-plugin-sofagent-daemon SKILL.md skill
kongfangxun/sofagent · engine/dsh-plugins/cordis-plugin-sofagent-daemon/SKILL.md · 7×24 巡检 + 健康监测 + webhook 推送(seam: non-seam:host-process)——桥接 @sofagent/daemon startCron——DSH(DeepSeek Harness)cordis…
v1.7.0 · audit A · 54 stars

Every file in kongfangxun/sofagent

Browse by kind, by grade A, or by owner.