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.
npx skills add willnie9/agent-skills --skill yunxiao-bug-fixgit clone --depth 1 https://github.com/willnie9/agent-skillsWrote 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.
[](https://agentmods.dev/skills/willnie9/agent-skills/yunxiao-bug-fix)<a href="https://agentmods.dev/skills/willnie9/agent-skills/yunxiao-bug-fix"><img src="https://agentmods.dev/badge/skills/willnie9/agent-skills/yunxiao-bug-fix/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.
<a href="https://agentmods.dev/skills/willnie9/agent-skills/yunxiao-bug-fix"><img src="https://agentmods.dev/badge/skills/willnie9/agent-skills/yunxiao-bug-fix.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00075 | $0.07894 |
| Opus 5 | $0.00037 | $0.03947 |
| Sonnet 5 | $0.00015 | $0.01579 |
| Haiku 4.5 | $0.00007 | $0.00789 |
Grade A, and why
yunxiao-bug-fix scanned grade A with 1 finding 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 10d ago.
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.
Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
4. **截图附件必须 `curl` 下载到本地查看**,即使文字描述已清晰。 How it starts
The opening of the file, as written. The whole thing — 712 lines — stays where its author put it; the contents beside it link to each section on GitHub.
云效 Bug 修复标准流程(SOP)
⭐ Auto Mode(无确认全自动模式)
触发:用户输入含任一关键词 → 全跑 / 一气呵成 / auto / yolo / 别问 / 自动,或一次贴多个 bug 链接说"一起改"。
单 bug Auto 流程
用户给单链接 + auto
↓
Step 1 定位 (单链接,无需搜索)
↓
Step 2 跳过(不展示等用户)
↓
Step 3 抓上下文 (4 子动作必做,铁律 3/4)
↓
Step 4 代码定位
↓
Step 5 跳过出方案(直接改)
↓
Step 6 改代码
↓
Step 7 浏览器验证 (auto 强制选 C 路径)
↓
┌───────────────────────────┴──────────────────┐
▼ ▼
verdict=pass(验证通过) verdict=fail(验证失败)
↓ ↓
Step 8 自动回写云效: ⚠️ 停下来等用户确认
① update_work_item → 开发完成 问用户:
② create_work_item_comment → 评论 "验证未通过(原因: 没数据/没详情/...)
③ create_workitem_attachment → 截图附件 要不要仍然回写云效?
A) 仍回写(我知道)
B) 不回写,我自己看代码
C) 重新跑验证"
↓ ↓
Step 9 落 bug-fix-report.json (verdict=pass) Step 9 落 bug-fix-report (verdict=fail/warn)
单 bug Auto 自动决策表
| 步骤 | 默认行为(等用户) | Auto 行为 |
|---|---|---|
| Step 1 定位 Bug | 列候选等用户选编号 | 单链接直接进;搜索结果按优先级取第一 |
| Step 2 展示等"开始修复" | ⚠️ 铁律 2 必停 | 跳过 |
| Step 3 抓上下文(4 子动作) | 全做 | 全做(铁律 3/4 仍生效) |
| Step 4 代码定位 | 同 | 同 |
| Step 5 出方案等"确认" | ⚠️ 铁律 5 必停 | 跳过 |
| Step 6 改代码 | 同 | 同 |
| Step 7 弹 A/B/C/D | ⚠️ 铁律 6 必停 | 强制走 C(验证+回写) |
| Step 8 浏览器验证 | 选 A/C 跑 | 跑(必跑) |
| Step 9 回写云效 | verdict 决定 | pass → 自动回写状态+评论+截图附件;fail → 停下问用户回写还是不回写 |
| Step 10 落 report | 同 | 同 |
多 bug Auto 流程(批处理,不是串行单 bug)
用户给 N 个链接 / "auto 把今天 5 个 bug 都改了"
↓
Phase 1: 批量改代码(N 个 bug 依次走 Step 1-6,只到改完代码不验证)
↓ (中间产物: 每个 bug 落 .claude/results/bug-fix-<id>/code-changed.json)
↓
Phase 2: 批量验证(每个 bug 用 MCP playwright 操作浏览器,按 §9.1 流程)
for bug in bugs:
用 MCP browser_* 走通验证步骤,落截图到 runtime/screenshots/<bug-id>.png
↓ (产物: .claude/skills/playwright-skill/runtime/screenshots/<bug-id>.png × N)
↓
Phase 3: 验证成功的统一回写云效(并行调 MCP,无 git 冲突所以可并行)
for bug in passed:
update_work_item({ status: '开发完成' })
create_work_item_comment({ ... })
create_workitem_attachment({ ... 截图 ... })
↓
Phase 4: 验证失败的停下问用户
"<BUG-1>|<BUG-2>|<BUG-3> 验证未通过(原因 X),要回写吗? Y/N"
用户决定
↓
最终汇总报告: ✅ 4/5 通过自动回写 / ⚠️ 1/5 待用户决定
What ships with it
2 files 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.
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.
- 10d ago First seen · 712 lines · 75 tokens per session scan A 98e2d4f84875
yunxiao-bug-fix is a skill published in the GitHub repository willnie9/agent-skills (2 stars, last pushed 2mo ago), licensed MIT. It adds 75 tokens to every session and 7,894 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
browser-edge-cases
SOP for debugging browser automation failures on complex websites. Use when browser tools fail on specific sites like LinkedIn, Twitter/X, SPAs, or sites with Shadow DOM.
triage-issue
Analyze a GitHub issue, verify claims against the codebase, and close invalid issues with a technical response.
diag-harness
Kernel-version skew check (ADR-027). Reports manifest surface + manifest kernel + installed kernel + verdict (match/patch-diff/minor-diff/major-diff). Exits 1 on minor/major skew with a copy-pasteable npm install @metaharness/[email protected] next step. Exits 2 if no .harness/manifest.json at path.
kn-debug
Use when debugging errors, test failures, build issues, or blocked tasks — structured triage to fix to learn.
config-validator
Validate AIWG configuration files and project setup for correctness and completeness.
cloudflare-workers-observability
Cloudflare Workers observability with logging, Analytics Engine, Tail Workers, metrics, and alerting. Use for monitoring, debugging, tracing, or encountering log parsing, metric aggregation, alert configuration errors.