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 agentmods add skills/deeplink-org/probing/watchdog_timeoutnpx skills add DeepLink-org/probing --skill watchdog_timeoutgit 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/watchdog_timeout)<a href="https://agentmods.dev/skills/deeplink-org/probing/watchdog_timeout"><img src="https://agentmods.dev/badge/skills/deeplink-org/probing/watchdog_timeout.svg" alt="Measured on agentmods" 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 | $0.00020 | $0.00503 |
| Opus 5 | $0.00010 | $0.00251 |
| Sonnet 5 | $0.00004 | $0.00101 |
| Haiku 4.5 | $0.00002 | $0.00050 |
Grade A, and why
watchdog_timeout 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 4d 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
Watchdog timeout diagnosis
用于 PyTorch NCCL watchdog timeout、collective deadlock、rank 间 collective 顺序不一致等场景。
这个 skill 会先触发 PyTorch Flight Recorder snapshot，并写入：
python.torch_nccl_flight_recordpython.torch_nccl_pg_status
然后按 pg_id + collective_seq_id 对齐各 rank，检查 missing rank、collective type mismatch、shape/dtype mismatch、state mismatch。
Prerequisites
- PyTorch >= 2.5
- 目标进程内存在
torch._C._distributed_c10d._dump_nccl_trace - 最好提前设置
TORCH_NCCL_TRACE_BUFFER_SIZE=2000 - timeout 自动落盘可额外设置
TORCH_NCCL_DUMP_ON_TIMEOUT=1
Reading results
ranks_seen < expected_world_size：某些 rank 没有进入同一个 collective seq，常见于 CPU-side hang 或分支发散。op_min != op_max：同一个 seq 上不同 rank 发起了不同 collective。input_sizes_min != input_sizes_max或 dtype 不同：collective 参数不一致。state_min != state_max：某些 rank 未完成、未开始或处于不同状态。
Related skills
- 已进入 NCCL 但想看 culprit/victim →
nccl_culprit_victim - 训练仍存活但无进展 →
training_hang - 粗粒度通信慢 →
comm_bottleneck
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.
- 4d ago First seen · 47 lines · 20 tokens per session scan A c9be725d5eec
watchdog_timeout is a skill published in the GitHub repository DeepLink-org/probing (11 stars, last pushed 5d ago), licensed Apache-2.0. It adds 20 tokens to every session and 503 once invoked, about $0.0001 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
release-validation
Use this skill when validating a MeshLLM release candidate or current HEAD against the last GitHub release, assembling the canonical feature/fix/modification inventory, testing locally built release bundles on user-approved real hosts and private meshes, deciding release readiness, or producing a formal…
hf-gguf-quant-jobs
Use when creating, monitoring, validating, or documenting low-memory Hugging Face Jobs or local runs that quantize split BF16/FP16 GGUF model repos into custom quant GGUF repos with skippy-quantize.
plugin-web-ui-extension
Use this skill when maintaining the plugin web UI projection contract, docs, exemplar coverage, or recovery flow for mesh-llm plugin web UI work.
skippy-bench
Use this skill when running benchmark orchestration, local single-stage or split benchmarks, benchmark report flow, or performance-oriented skippy runtime checks.
deploy-windows
Use this skill when installing, deploying, launching, serving, or troubleshooting mesh-llm on a Windows machine — PowerShell install via install.ps1, flavor selection (CUDA/ROCm/Vulkan/CPU), source builds, the contrib helper scripts, and verifying it serves.
hf-bf16-gguf-conversion-jobs
Use when converting Hugging Face SafeTensors checkpoints into split BF16 GGUF model repos with skippy-quantize on Hugging Face Jobs or a local machine, then publishing the artifact to Hugging Face.