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 AbyssCN/oh-my-dag --skill omd-debuggit clone --depth 1 https://github.com/AbyssCN/oh-my-dagWrote 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/abysscn/oh-my-dag/omd-debug)<a href="https://agentmods.dev/skills/abysscn/oh-my-dag/omd-debug"><img src="https://agentmods.dev/badge/skills/abysscn/oh-my-dag/omd-debug.svg" alt="Measured on agentmods" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00084 | $0.01180 |
| Opus 5 | $0.00042 | $0.00590 |
| Sonnet 5 | $0.00017 | $0.00236 |
| Haiku 4.5 | $0.00008 | $0.00118 |
Grade A, and why
omd-debug 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 8d 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.
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.
What it actually says
/omd-debug — 系统化根因调试
铁律:无根因不修。 修症状 = 打地鼠。找到根因再动手,诊断阶段禁提修复方案。
方法论 skill:单线程串行用 runtime 原生 Grep/Read/git 走完 8 阶段;假设验证需并发多路时派 omd dag_debug(一假设一 leaf + skeptic 跨模型证伪 + codegraph 锁范围 + 三振纪律,见 Phase 5)。
Phase 1 — 历史搜索(先查后调)
同一消息并行发:git log --oneline -20 -- <相关文件>(近期变更)+ Grep 既有 docs/plan/·NOTES·error 记录搜错误关键词(有先例走 /omd-recall 先召回)。命中已知修复 → 直接引用,跳到修复。
Phase 2 — 复现
确定性触发 + 存证据(错误信息/堆栈/复现步骤)。信息不足 → 逐个追问(一次一问)。不能稳定复现 → 收集更多证据再继续,别猜。
Phase 3 — Scope Lock
从症状追到最窄受影响目录,声明范围;此后 Edit 只在范围内,越界先问 owner。危险信号立即停:「顺手修个相关的」(不顺手,记 follow-up)、「既然在改不如重构」(不重构,本 bug 修完再评估)。
Phase 4 — 模式匹配
先 step-back 自问:这类症状(数据不一致/超时/类型错/竞态/权限泄漏/schema 漂移)在本技术栈通常由哪层引起?建宏观诊断框架再缩范围,别直接跳到最"像"的模式。同文件反复修 bug = 架构问题,不是巧合。
Phase 5 — 假设验证(最多 3 个)
每假设三步推导:观察(引 file:line 我看到什么)→ 推论(若成立还应看到什么)→ 验证(设计能区分成立/不成立的实验)。追踪表标 CONFIRMED/REJECTED。先验证再修(可疑根因处加临时 log/断言,跑复现看证据是否吻合)。三振出局:3 个假设全败 → STOP 问 owner(A 继续有新假设 / B 升级人工 / C 埋点观察)。危险信号:还没追数据流就提修复 = 在猜。
并发扇出 → 派
dag_debug:假设多、想并行验(一假设一 leaf + skeptic 证伪 + judge 收敛)而非单线程串行时,派 omddag_debug(failure 必填,repro/oracleCmd可选)。它把本阶段的假设验证 DAG 化,同守本 skill 铁律(无根因不修 / 默认只提议不改文件 / 三振停升 owner)。复现拿 red + codegraph 锁范围由引擎代跑;finding≠ground truth,回来仍你终裁。
Phase 6 — 修复
根因确认后最小改动:最少文件、最少行,不顺手重构相邻代码。写回归测试(不修 FAIL、修后 PASS),跑测试套件贴输出。改 >5 文件 → 问 owner(bugfix 爆炸半径偏大:继续 / 拆分 / 重想)。
Phase 7 — 验证闭环
复现原始场景确认已修 + 跑变更相关验证。不说"应该能修好",验证并证明。
Phase 8 — 报告 + 扫同类
报告:症状 / 根因(为什么)/ 模式 / 修复(file:line)/ 证据 / 回归测试 / 状态(DONE · DONE_WITH_CONCERNS · BLOCKED)。修完扫一片:用根因反推同类——Grep 同 pattern(根因是"缺 scope 过滤" → 全仓找所有 db 访问点;"未校验入参" → 找所有 untrusted 入口)。>5 处别一次全修,修 1-2 关键,其余记 NOTES 留 follow-up。
与既有 skill 的边界
- omd-debug = 某失败的根因调查。审一批 diff 找 bug → /omd-review;扫安全漏洞 → /omd-audit;找过度工程 → /omd-slim。修复方案涉及大改前可用 /omd-grill 审。
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.
- 8d ago First seen · 40 lines · 84 tokens per session scan A c823da5e3153
omd-debug is a skill published in the GitHub repository AbyssCN/oh-my-dag (39 stars, last pushed 3d ago), licensed MIT. It adds 84 tokens to every session and 1,180 once invoked, about $0.0004 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 2026-08-30.
Other skills, from other repositories
optimization
Use when improving performance, latency, throughput, memory usage, or general efficiency. Start by defining target metrics, measuring comprehensively, attributing bottlenecks, validating with static analysis, and prioritizing macro-optimizations before micro-optimizations.
smiles-validation
Strict SMILES validation, structural comparison, and modification verification. Catches invalid LLM-generated molecules.
neuron-debugger
Debug and monitor Neuron AI applications with Inspector APM, event observability, logging, and performance analysis. Use this skill whenever the user mentions debugging, monitoring, observability, performance analysis, tracing, Inspector, or needs to understand why an agent is behaving a certain way. Also trigger for…
data-fetching
Use when implementing or debugging ANY network request, API call, or data fetching. Covers fetch API, axios, React Query, SWR, error handling, caching strategies, offline support.
gini-bug-report
File a locally-captured, already-redacted Gini crash report as a GitHub issue, with the user's consent. Reads the pending crash queue and delegates the actual filing to the github-issues skill.
pi-observability-local-dev
Use when developing or reviewing packages/pi-observability dashboard/server changes locally, especially rebuilding dist, restarting the 43190 server, validating UI behavior, or keeping the local dashboard in sync with source edits.