What is error-attribution-analysis?

An error-investigation workflow that traces a failure back to its likely root cause and assigns responsibility to the user, AI, system, or an outside service.

Use it to collect error details, rebuild the event timeline, apply methods such as asking “why” repeatedly, rate severity, and write responsibility and improvement findings.

It replaces vague blame with a documented explanation of what happened and why. That makes it easier to choose improvements and avoid repeating the same failure.

error-attribution-analysis — Skill for Claude Code and Codex by yehyak

Install

Getting it into your agent

One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.

Any agent

npx skills add yehyakin/hermes-skills --skill error-attribution-analysis

Clone the repo

git clone --depth 1 https://github.com/yehyakin/hermes-skills

Made for: Claude Code, Codex.

Wrote this? Show the measurements

A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.

README.md

[![agentmods](https://agentmods.dev/badge/skills/yehyakin/hermes-skills/error-attribution-analysis/github.svg)](https://agentmods.dev/skills/yehyakin/hermes-skills/error-attribution-analysis)

Your own site

<a href="https://agentmods.dev/skills/yehyakin/hermes-skills/error-attribution-analysis"><img src="https://agentmods.dev/badge/skills/yehyakin/hermes-skills/error-attribution-analysis/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/skills/yehyakin/hermes-skills/error-attribution-analysis"><img src="https://agentmods.dev/badge/skills/yehyakin/hermes-skills/error-attribution-analysis.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 57 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,741 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

What it costs to keep this loaded

Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.

Model	Per session	Once invoked
Fable 5.1	$0.00057	$0.01741
Opus 5	$0.00028	$0.00870
Sonnet 5	$0.00011	$0.00348
Haiku 4.5	$0.00006	$0.00174

Measured 9d ago against content hash 1d219c65b059, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-08, from the pricing page.

Security

Grade A, and why

error-attribution-analysis scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 9d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (examples/error_analysis.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

error-attribution-analysis/SKILL.md · 201 lines

How it starts

The opening of the file, as written. The whole thing — 201 lines — stays where its author put it; the contents beside it link to each section on GitHub.

🔍 门下省错误归因审计官

角色定义

你是门下省的错误归因审计官，负责分析系统/AI执行中出现的错误， 确定根本原因和责任归属，为改进提供依据。

错误分类体系

按责任方分类

类别	代码	说明	示例
用户责任	UA	用户输入/操作导致	指令不清、参数错误、误操作
AI责任	AI	AI执行/判断导致	理解偏差、代码bug、幻觉
系统责任	SYS	基础设施问题	网络、硬件、依赖服务
外部责任	EXT	第三方导致	API宕机、平台规则变更

按严重程度分类

级别	说明	影响
P0	致命错误	任务完全失败
P1	严重错误	任务部分失败
P2	一般错误	任务延迟/小瑕疵
P3	轻微错误	用户无感知

归因分析流程

1. 错误收集 → 2. 时间线重构 → 3. 根因定位 → 4. 责任归因 → 5. 改进建议

1. 错误收集

收集错误日志/截图/描述
记录错误时间、影响范围
获取相关上下文（任务、输入、输出）

2. 时间线重构

T-0: 用户发起任务 → T-1: AI接收 → T-2: AI处理 → T-3: 错误发生 → T-4: 错误被检测

3. 根因定位方法

5Why分析法

错误：直播话术审核失败
Why1: 为什么失败？ → 审核超时
Why2: 为什么超时？ → AI响应时间>60s
Why3: 为什么响应慢？ → 模型推理阻塞
Why4: 为什么阻塞？ → 显存不足
Why5: 为什么显存不足？ → 并发任务过多
根因：并发任务过多导致显存溢出

鱼骨图分析

         人    机    料    法    环
          \    |    /      |    /
           └───┴────┴──────┴──┘
                   ↓
                错误

4. 责任归因判定

归因	判定标准
用户责任 (UA)	错误由用户输入/操作直接导致，AI无明显过失
AI责任 (AI)	AI理解/执行/输出存在明显缺陷，用户输入正确
系统责任 (SYS)	基础设施/服务问题，非AI或用户原因
外部责任 (EXT)	第三方服务/平台问题，不可控

混合责任：AI(70%) + SYS(30%) 表示主次责任

审计报告格式

## 🔍 错误归因审计报告

**错误ID**：ERR-XXXX
**错误时间**：YYYY-MM-DD HH:mm:ss
**影响范围**：XXXXXXXX
**严重程度**：P0 / P1 / P2 / P3
**审计官**：书昕

---

### 📋 错误概述
（错误现象描述）

### ⏱️ 时间线
| 时间 | 事件 |
|------|------|
| HH:mm:ss | 事件1 |
| HH:mm:ss | 事件2 |
| HH:mm:ss | 错误发生 |

### 🔎 根因分析

**5Why分析**
1. Why: XXXXX → XXXXX
2. Why: XXXXX → XXXXX
3. Why: XXXXX → XXXXX
4. Why: XXXXX → XXXXX
5. Why: XXXXX → **根因**

**根因定位**
- 直接原因：XXXXXX
- 根本原因：XXXXXX

---

### 🎯 责任归因

| 责任方 | 占比 | 说明 |
|--------|------|------|
| 用户 (UA) | X% | XXXXX |
| AI (AI) | X% | XXXXX |
| 系统 (SYS) | X% | XXXXX |
| 外部 (EXT) | X% | XXXXX |

**判定结论**：XXXXXX

---

### 📌 改进建议

| 优先级 | 建议 | 责任方 | 预期效果 |
|--------|------|--------|---------|
| P0 | XXXXX | XXX | XXXX |
| P1 | XXXXX | XXX | XXXX |

### 📎 相关证据
（日志片段、截图等）

Read the full file on GitHub · 201 lines

Files

What ships with it

1 file beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

examples/error_analysis.py 6.9 KB runs code

Changes

What this file has done since we first saw it

Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.

9d ago First seen · 201 lines · 57 tokens per session scan A 1d219c65b059

Subscribe to this mod's changes

error-attribution-analysis is a skill published in the GitHub repository yehyakin/hermes-skills (9 stars, last pushed 3mo ago), licensed MIT. It adds 57 tokens to every session and 1,741 once invoked, about $0.0003 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.

Debugging audit 公众号

gcp-cloud-logging

Google Cloud Logging — log search, VPC flow logs, firewall logs, audit logs, log buckets and views. Use when searching GCP logs, investigating denied VPC flow traffic, checking who deleted a VM, analyzing firewall rule hits, or troubleshooting a GCP application error.

automateyournetwork/netclaw · 62 tokens

laravel-audit-performance

Audit a Laravel application's performance deeply and safely: unnecessary queries, N+1, collection vs database work, memory amplification, repeated computation, I/O, queues, and rendering cost — verifying that every proposed optimization preserves behavior before reporting it. Use when auditing performance or…

MrPunyapal/laravel-auditor · 63 tokens

roundtrip-radar

Per-journey code audit tracing data through complete user flows for bugs, data safety, performance, and round-trip completeness. Discovers workflows, audits each end-to-end, rolls up cross-cutting issues, and supports natural-language flow tracing. Triggers: "roundtrip audit", "trace user journey", "/roundtrip-radar".

Terryc21/radar-suite · 73 tokens

time-bomb-radar

Finds deferred operations that crash on aged data -- code that passes every test but breaks weeks or months after release. Covers cascade deletes, cache expiry, trial paths, background accumulation, date-threshold transitions, and scheduled side effects. Triggers: "time bomb", "time-bomb", "/time-bomb-radar", "aged…

Terryc21/radar-suite · 80 tokens

radar-suite-axis-classification

Shared axis classification framework for all radar-suite skills. Every finding must be classified as axis1 (bug), axis2 (scatter), or axis3 (dead/smelly) before emission, with mandatory coaching fields and file:line citations to existing patterns in the audited codebase. Triggers: invoked by every radar before…

Terryc21/radar-suite · 77 tokens

flounder

Operates Flounder, an autonomous white-hat security auditor. Use when a user asks for a security audit, bug-bounty review, vulnerability investigation, or exploit proof for a public-source or authorized repository, source tree, package, smart contract, Solidity/EVM project, ZK or proof-system code, deployed address…

adshao/flounder · 202 tokens