What is ascend-operator-debug?

A debugging workflow that reduces an Ascend model failure to one operator call and tests it against a reference implementation. An operator is a low-level computation such as a tensor transformation or mathematical operation.

Use it to investigate Ascend operator crashes and dtype, shape, layout, or execution-mode issues, create a focused reproducer, record case results, and add the smallest failing case as a regression test.

A failure in a complete model does not prove that one operator is responsible; isolating the call reveals whether the issue is a crash, unsupported input, numerical mismatch, or integration problem.

How maoxx241/vllm-ascend-workspace configures Codex

Reuse

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.

Copy the file

curl -O https://raw.githubusercontent.com/maoxx241/vllm-ascend-workspace/main/.agents/skills/ascend-operator-debug/SKILL.md

Clone the repo

git clone --depth 1 https://github.com/maoxx241/vllm-ascend-workspace

Made for: Codex.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/maoxx241/vllm-ascend-workspace/ascend-operator-debug/github.svg)](https://agentmods.dev/skills/maoxx241/vllm-ascend-workspace/ascend-operator-debug)

Your own site

<a href="https://agentmods.dev/skills/maoxx241/vllm-ascend-workspace/ascend-operator-debug"><img src="https://agentmods.dev/badge/skills/maoxx241/vllm-ascend-workspace/ascend-operator-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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/maoxx241/vllm-ascend-workspace/ascend-operator-debug"><img src="https://agentmods.dev/badge/skills/maoxx241/vllm-ascend-workspace/ascend-operator-debug.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 91 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 589 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe. Third-party audits

NVIDIA SkillSpector pass 7 Sept 2026

How audits are shown

Origin original No closer match found in the catalogue.

Token cost

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	$0.00091	$0.00589
Opus 5.5	$0.00036	$0.00236
Sonnet 5.5	$0.00018	$0.00118
Haiku 4.5	$0.00009	$0.00059

Measured 2mo ago against content hash 9bf94487695a, method: parsed. Prices are Anthropic first-party input rates as of 2026-10-07, from the pricing page.

Security

Grade A, and why

ascend-operator-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 2mo ago.

The scan reads SKILL.md. This mod also ships 2 executable files (scripts/operator_debug.py, tests/test_operator_debug.py), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

.agents/skills/ascend-operator-debug/SKILL.md · 58 lines

How it starts

The opening of the file, as written. The whole thing — 58 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Ascend Operator Debug

Turn a suspected operator failure into a portable reproducer and an explicit case matrix. A model-level symptom is not an operator bug until the isolated call reproduces it.

Workflow

Capture the failing call's operator name, arguments, input metadata, execution mode, environment, and source stack without copying full tensors by default.
Define a trusted reference implementation and tolerances before comparing.
Create an explicit case matrix with scripts/operator_debug.py plan.
Run the generated cases on a remote Ascend environment. Change one dimension at a time: dtype, shape, layout, mode, or operator option.
Record each normalized result with record.
Run analyze to separate numerical mismatch, crash, unsupported combination, missing evidence, and operator-pass/integration-fail outcomes.
Add the smallest failing case as a regression test, then rerun the original model integration after the operator fix.

Entry point

scripts/operator_debug.py provides:

plan: validate the explicit case matrix and create the evidence layout;
record: accept one result for a planned case without overwriting evidence;
analyze: summarize failure axes and create a Run Manifest-linked report.

Read only the reference needed for the current phase:

Boundaries

This skill begins after evidence identifies one operator boundary or the user explicitly supplies an operator reproducer.
Whole-model eager-versus-graph localization belongs to vllm-ascend-graph-debug; hand off only after one operator call is isolated.
Rank-dependent and collective failures belong to vllm-ascend-distributed-debug.
Model-level throughput regressions belong to performance workflows. An optional operator timing value here is only supporting evidence for the isolated case.

Rules

Read the full file on GitHub · 58 lines

Files

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.

Changes

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.

2mo ago First seen · 58 lines · 91 tokens per session scan A 9bf94487695a

Subscribe to this mod's changes

ascend-operator-debug is a skill published in the GitHub repository maoxx241/vllm-ascend-workspace (36 stars, last pushed 2mo ago), licensed MIT. It adds 91 tokens to every session and 589 once invoked, about $0.0004 per session on Opus 5.5. 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.

Debugging

ascend-operator-debug next to 6 skills from other repositories
Mod	Type	Stars	Last push	Licence	Scan	Tokens	Works with
ascend-operator-debugmaoxx241/vllm-ascend-workspace	Skill	36	2mo ago	MIT	A	91	Codex
systematic-debuggingopen-metadata/OpenMetadata	Skill	16k	yesterday	Apache-2.0	A	37	Claude Code
log-error-digestzebbern/claude-code-guide	Skill	4.7k	3d ago	MIT	A	71	Claude Code, Codex
cwv-optimizeradobe/skills	Skill	319	yesterday	Apache-2.0	A	112	Claude Code
ci-fixersipyourdrink-ltd/bernstein	Skill	1.2k	yesterday	Apache-2.0	A	15	Claude Code, Codex
incident-response-incident-responsermyndharis/antigravity-skills	Skill	1.9k	23d ago	MIT	A	36	Claude Code
triaging-web-analytics-supportPostHog/posthog	Skill	41k	yesterday	none	A	151	Claude Code, Codex

vllm-ascend-workspace: Skill for Codex