semantic-audit SKILL.md skillA

semantic-audit from echoyue-lp/echo-agent is an agent skill: a SKILL.md folder an AI agent loads when a task calls for it.

Install it, pinned
mdr add echoyue-lp/echo-agent/semantic-audit@git:20260908.7e2cb50

Pins git:20260908.7e2cb50 in mdr.lock. Pin the fingerprint instead.

Audit grade
A, 17 of 17 checks passed
Kind
SKILL.md skill
Repository
echoyue-lp/echo-agent
Path
.agents/skills/semantic-audit/SKILL.md
Versions recorded
1
Latest version
git:20260908.7e2cb50, committed Sep 8, 2026
Size
2,333 bytes
Repository stars
89
License
MIT

针对指定语义边界和风险视角进行有界、可反证的质量审查,记录故障假设、实现路径、证据、问题记录和残余风险; 仅在已有“边界 × 风险视角 × 源码版本”单元及进入原因时使用,不负责普通差异分析或一般代码审查。.

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

# 语义质量审查

审查记录“在某个源码版本上检查过哪些故障假设”,不记录绝对安全结论。审查默认只读,发现问题后
把修复交回正常研发流程。

## 调度

- 开始前调用 `semantic-contract`,取得审查和问题记录字段;合同、基线或边界不存在时停止。
- 从当前待审集合优先领取失效的高风险单元、残余风险、新问题记录或事故相关单元。
- 一轮最多处理 3 个审查包;每包 1~3 个“语义边界 × 风险视角 × 源码版本”单元。
- 只有同一边界、同一源码版本、实现路径和证据获取方式实质重叠时才合包,不为凑数拼接弱相关视角。
- 审查 Subagent 之间不共享中间结论;需要独立复核时明确记录原因。复用既有 Subagent/TaskRuntime 准入,
  不新建审查专用并发模型。
- 优先派发 `semantic-risk-reviewer`。宿主不能发现专用角色时,改用现有只读 `reviewer`,并在派发任务中完整提供
  仓库绝对路径、源码版本、边界、风险视角、进入原因、相关对象、排除项和回报合同。无法取得独立上下文时,
  由主 Agent 继续只读检查并把“未完成独立复核”列入证据缺口。

## 每个单元必须回答

1. 明确的故障假设和触发条件;
2. 实际入口、状态权威、失败/并发/恢复路径和外部副作用;
3. 支持或反驳结论的源码、契约、测试或脱敏运行证据;
4. 问题记录、残余风险和未检查项。

没有发现问题记录也只能记录已检查假设和残余风险,不能写 `clean`、`safe` 或“没有缺陷”。泛泛的模块评价
和风险清单不能把审查标记为 `examined`。不要在本 Skill 内修改业务代码或关闭问题记录。

需要确认真实期望或接受风险时调用 `semantic-decide`;其余结果调用 `semantic-verify` 校验本轮写入。

Read the whole file at its exact version.

How to install

The exact contents, by fingerprint
mdr add echoyue-lp/echo-agent/semantic-audit@sha256:ccc09104b4177171

The line at the top of this page pins the label git:20260908.7e2cb50, 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_abvioqyiivxrcusk.svg)](https://markdownregistry.com/a/art_abvioqyiivxrcusk)

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

Versions

The registry has recorded 1 version of this file, committed on Sep 8, 2026.

versioncommittedcommitsizeaudit
git:20260908.7e2cb50 latest2026-09-08 7e2cb50 2,333 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 (2333 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

echoyue-lp/echo-agent · 89 stars · license MIT · pushed 2026-10-31 · branch main

API

GET https://markdownregistry.com/api/v1/artifacts/art_abvioqyiivxrcusk
GET https://markdownregistry.com/api/v1/resolve?ref=echoyue-lp/echo-agent/semantic-audit
GET https://markdownregistry.com/api/v1/blob/ccc09104b417717137cebcba68575654a9a6e1527dda637610ba5e850fc2ae5f

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 echoyue-lp/echo-agent

coding-preflight SKILL.md skill
echoyue-lp/echo-agent · .agents/skills/coding-preflight/SKILL.md · 在 echo-agent 开始缺陷修复、功能开发、重构、契约或样式改动前,先完成复用搜索、架构分层、状态权威、 修改边界和验证范围预检;当新需求可能导致重复代码、相近概念、公共接口漂移或风格分裂时使用,不负责首次全量语义建库。
git:20260908.7e2cb50 · audit A · 89 stars
semantic-contract SKILL.md skill
echoyue-lp/echo-agent · .agents/skills/semantic-contract/SKILL.md · 提供语义材料版本 1 的唯一对象、引用、状态、风险视角和结构校验合同;仅在其它语义 Skill 要读写语义对象, 或用户要求检查语义材料格式与合同漂移时使用,不负责普通代码审查。
git:20260908.7e2cb50 · audit A · 89 stars
semantic-decide SKILL.md skill
echoyue-lp/echo-agent · .agents/skills/semantic-decide/SKILL.md · 把真实产品预期、语义冲突或残余风险整理成一个需要人的明确裁决单元; 仅在已有具体行为承诺、规则或问题记录无法由证据确定时使用,不负责普通设计讨论或一般代码审查。
git:20260908.7e2cb50 · audit A · 89 stars
semantic-diff SKILL.md skill
echoyue-lp/echo-agent · .agents/skills/semantic-diff/SKILL.md · 分析代码、配置、契约或工作树差异对既有语义基线的影响,计算影响闭包并标记需要重新审查的风险视角; 仅在已有实际差异且存在语义基线时使用,不负责首次建库、编码前预检或泛化代码审查。
git:20260908.7e2cb50 · audit A · 89 stars
semantic-discover SKILL.md skill
echoyue-lp/echo-agent · .agents/skills/semantic-discover/SKILL.md · 建立或补全仓库的语义行为基线,识别能力边界、状态权威、重要入口、证据缺口和未知区; 仅在首次建立、扩大或修复 `semantic/` 基线覆盖时使用,不负责普通新需求预检、已有差异分析或代码审查。
git:20260908.7e2cb50 · audit A · 89 stars
semantic-verify SKILL.md skill
echoyue-lp/echo-agent · .agents/skills/semantic-verify/SKILL.md · 校验语义材料的结构、引用、源码快照、新鲜度和解决条件,并在静态事实不足时运行最小必要验证; 仅在已有 `semantic/` 材料或问题记录解决候选时使用,不替代普通代码完成验证或仓库合并门禁。
git:20260908.7e2cb50 · audit A · 89 stars
AGENTS.md AGENTS.md file
echoyue-lp/echo-agent · AGENTS.md
git:20261023.5172645 · audit A · 89 stars
code-review SKILL.md skill
echoyue-lp/echo-agent · echo-agent-learning/examples/demo_skills/code-review/SKILL.md · Professional code review skill: identify defects, security risks, and best practice violations. Use when asked to…
v2.0.0 · audit A · 89 stars
data-analyst SKILL.md skill
echoyue-lp/echo-agent · echo-agent-learning/examples/demo_skills/data-analyst/SKILL.md · Data analysis skill: statistical analysis, data interpretation, and visualization recommendations. Use when asked to…
v2.0.0 · audit A · 89 stars
project-stats SKILL.md skill
echoyue-lp/echo-agent · echo-agent-learning/examples/demo_skills/project-stats/SKILL.md · Analyze project structure, count lines of code by language, find TODOs/FIXMEs, and generate project health summaries…
v2.0.0 · audit A · 89 stars
python-linter SKILL.md skill
echoyue-lp/echo-agent · echo-agent-learning/examples/demo_skills/python-linter/SKILL.md · Lint and format Python files using ruff. Activates automatically when Python files are touched. Use to check code…
v1.0.0 · audit A · 89 stars
web-researcher SKILL.md skill
echoyue-lp/echo-agent · echo-agent-learning/examples/demo_skills/web-researcher/SKILL.md · Web research skill: systematically collect, verify, and synthesize information from multiple sources. Use when asked to…
v1.2.0 · audit A · 89 stars

Every file in echoyue-lp/echo-agent

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