Borrowing it
Nothing to install: this file belongs to maoxx241/vllm-ascend-workspace. Take a copy, put it at the same path in your own repository, and replace the rules that are about this project with yours.
curl -O https://raw.githubusercontent.com/maoxx241/vllm-ascend-workspace/main/.agents/skills/vllm-ascend-distributed-debug/SKILL.mdgit clone --depth 1 https://github.com/maoxx241/vllm-ascend-workspaceWrote 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/maoxx241/vllm-ascend-workspace/vllm-ascend-distributed-debug)<a href="https://agentmods.dev/skills/maoxx241/vllm-ascend-workspace/vllm-ascend-distributed-debug"><img src="https://agentmods.dev/badge/skills/maoxx241/vllm-ascend-workspace/vllm-ascend-distributed-debug/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/maoxx241/vllm-ascend-workspace/vllm-ascend-distributed-debug"><img src="https://agentmods.dev/badge/skills/maoxx241/vllm-ascend-workspace/vllm-ascend-distributed-debug.svg" alt="Reviewed on agentmods" width="80" 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 | $0.00096 | $0.00582 |
| Opus 5 | $0.00048 | $0.00291 |
| Sonnet 5 | $0.00019 | $0.00116 |
| Haiku 4.5 | $0.00010 | $0.00058 |
Grade A, and why
vllm-ascend-distributed-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 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.
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.
How it starts
The opening of the file, as written. The whole thing — 59 lines — stays where its author put it; the contents beside it link to each section on GitHub.
vLLM Ascend Distributed Debug
Build a falsifiable diagnosis from rank-aware evidence. Never infer a distributed root cause from one rank's log alone.
Workflow
- Create a case with
scripts/distributed_debug.py init. - Capture the exact failing topology, environment, process tree, endpoints, and reproduction command before changing parallelism.
- Add structured per-rank events with
ingest. Keep raw rank logs and stack dumps in the case directories created byinit. - Run
analyzeto check rank identity, group membership, endpoint collisions, collective order, missing participants, entered-without-exit stalls, and cross-rank phase divergence. - Form one or more falsifiable hypotheses from the report.
- Reduce one parallel dimension at a time. Record each reduced case separately.
- After a fix, rerun both the smallest reproducer and the original topology.
Entry point
scripts/distributed_debug.py provides:
init: validate the topology contract and create the complete evidence layout;ingest: validate and append normalized rank events;analyze: produce deterministic findings, per-rank last progress, and a Run Manifest-linked report.
Read only the reference needed for the current phase:
Boundaries
- This skill owns failures whose explanation requires comparing ranks, groups, nodes, or distributed endpoints.
- Correct outputs in eager mode with graph-only failure belong to
vllm-ascend-graph-debug. - A failure reduced to one operator call belongs to
ascend-operator-debug. - Kernel timing, throughput, and imbalance quantification belong to profiling or performance skills; do not load them until the distributed failure is stable and the user asks for that evidence.
Rules
- Preserve raw evidence; normalize into new files rather than rewriting logs.
- Treat missing ranks as missing evidence, not proof that those ranks crashed.
- Treat a collective mismatch as confirmed only when group, sequence, operation, and participating ranks are explicit.
- Redact secrets before storing environment snapshots.
- Keep cases under
.vaws-local/distributed-debug/.
What ships with it
6 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 · 59 lines · 96 tokens per session scan A 9c64352dd509
vllm-ascend-distributed-debug is a skill published in the GitHub repository maoxx241/vllm-ascend-workspace (36 stars, last pushed 6d ago), licensed MIT. It adds 96 tokens to every session and 582 once invoked, about $0.0005 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
debug-optimize-lcp
Guides debugging and optimizing Largest Contentful Paint (LCP) using Chrome DevTools MCP tools. Use this skill whenever the user asks about LCP performance, slow page loads, Core Web Vitals optimization, or wants to understand why their page's main content takes too long to appear. Also use when the user mentions…
systematic-debugging
Use when debugging a failing test, build error, or runtime issue that isn't immediately obvious. Guides a 4-phase root cause analysis instead of random fix attempts.
diagnose
Trace from a reproduced symptom to the source code that causes it. Pin the specific file and approximate line, rate confidence in the cause and clarity of the fix independently, and always propose a concrete fix.
repro-admin
Reproduce an EmDash admin UI bug. Attach a container, start the demo dev server, drive the admin with agent-browser using the dev-bypass session, and capture the reproduction as screenshots plus a replayable transcript.
log-error-digest
Analyze log files to troubleshoot errors, identify peak error periods, and produce error clustering, frequency statistics, and time distribution reports. Supports JSON, syslog, and Nginx formats with automatic detection. Use when a user uploads a .log file and asks to analyze errors, find patterns, debug issues, or…
byted-util-volcengine-detect-retry
An orchestration workflow for Volcengine Cloud Detect, a service that checks websites or network endpoints from test locations.