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 DeepLink-org/probing --skill sre_triagegit clone --depth 1 https://github.com/DeepLink-org/probingWrote 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/deeplink-org/probing/sre_triage)<a href="https://agentmods.dev/skills/deeplink-org/probing/sre_triage"><img src="https://agentmods.dev/badge/skills/deeplink-org/probing/sre_triage/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/deeplink-org/probing/sre_triage"><img src="https://agentmods.dev/badge/skills/deeplink-org/probing/sre_triage.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 | $0.00031 | $0.00565 |
| Opus 5 | $0.00015 | $0.00282 |
| Sonnet 5 | $0.00006 | $0.00113 |
| Haiku 4.5 | $0.00003 | $0.00056 |
Grade A, and why
sre_triage 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.
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
SRE triage
面向值班第一响应：把原本需要人工 ssh、grep、py-spy、看 NCCL/IB/GPU/内核日志的流程，尽量转成 probing 可执行证据链。
这个 skill 不追求一次性给出最终根因，而是自动回答：
- 集群/rank 是否都在线？
- 训练是否还在推进？
- 是否有 NCCL watchdog / collective desync 证据？
- 是否像 CPU hang、GPU/NCCL wait、网络拥塞、OOM/Xid？
- 下一步应该跑哪个更专门的 skill？
Reading results
incident_snapshot：当前有哪些表，决定能自动化到什么深度。cluster_nodes：rank/host 是否齐，是否有 dead/stale 节点。blocked_threads/kernel_wait_points：类似人工 py-spy/strace 的第一层替代。flight_recorder_alignment：自动化 Flight Recorder / fr_trace 的核心对齐判断。nccl_wait_summary：进入 NCCL 后的 culprit/victim 信号。rdma_network_hints/kernel_fault_hints：网络或硬件灰故障线索。
Related skills
- NCCL timeout 细查 →
watchdog_timeout - 已进入 NCCL wait →
nccl_culprit_victim - 训练卡住但不是 NCCL →
training_hang - 显存/OOM →
memory_leak/gpu_pressure
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.
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 · 47 lines · 31 tokens per session scan A 7ed4a22732a5
sre_triage is a skill published in the GitHub repository DeepLink-org/probing (11 stars, last pushed 3d ago), licensed Apache-2.0. It adds 31 tokens to every session and 565 once invoked, about $0.0002 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
k8s-incident-triage
Read-only Kubernetes incident triage using kubectl. Accepts natural language or structured input. Produces root-cause hypotheses, evidence, and next-step commands.
pager-triage
AI-powered incident triage for PagerDuty. List active incidents, deep-dive with timeline and alert correlation, check on-call schedules, acknowledge, resolve, and annotate — all from your agent. Read-only by default; write operations require explicit --confirm.
eng-runbook
A template for an engineering runbook, which is a practical guide engineers use during service operation and on-call incidents.
Incident Response Planner
Analyze an infrastructure incident alert and produce a structured response plan plus an executable on-call runbook. Use for incident, alert, on-call, and runbook requests.
incident-response
End-to-end incident workflow — find what's down, diagnose it, pause flapping monitors, and verify recovery.
Diagnose CrashLoopBackOff
Triage pods restarting repeatedly and identify the most likely root cause.