vllm-ascend-workspace: Skill for Codex

.agents/skills/vllm-ascend-pd-serving/SKILL.md

vllm-ascend-pd-serving is a skill for Codex from maoxx241/vllm-ascend-workspace. It costs 99 tokens per session (639 once invoked), scanned A, original, MIT.

A tool for planning, starting, checking, testing, and stopping a multi-session vLLM Ascend deployment split into prefill and decode services. vLLM serves language models, while prefill and decode are two stages of generating a response.

In plain words
What is it for?
Use it to operate PD-disaggregated vLLM deployments with NIXL, Mooncake, or another KV connector, when the required sessions and proxy already exist.
Why use it?
It coordinates the service order, connector settings, proxy health, rollback, and end-to-end smoke tests across separate sessions.

Skill for Codex

Written for Codex: agents/openai.yaml present. Also seen: installed under .agents/ (shared by several agents).

This is maoxx241/vllm-ascend-workspace's own configuration. It tells Codex how to work on vllm-ascend-workspace itself, so it is not a mod to install elsewhere. Copy it as a starting point and replace the rules that are about this project. Everything vllm-ascend-workspace configures →

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/vllm-ascend-pd-serving/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.

agentmods badge for vllm-ascend-pd-serving

README.md
[![agentmods](https://agentmods.dev/badge/skills/maoxx241/vllm-ascend-workspace/vllm-ascend-pd-serving/github.svg)](https://agentmods.dev/skills/maoxx241/vllm-ascend-workspace/vllm-ascend-pd-serving)
Your own site
<a href="https://agentmods.dev/skills/maoxx241/vllm-ascend-workspace/vllm-ascend-pd-serving"><img src="https://agentmods.dev/badge/skills/maoxx241/vllm-ascend-workspace/vllm-ascend-pd-serving/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.

agentmods 80×15 button for vllm-ascend-pd-serving

Your own site · 80×15
<a href="https://agentmods.dev/skills/maoxx241/vllm-ascend-workspace/vllm-ascend-pd-serving"><img src="https://agentmods.dev/badge/skills/maoxx241/vllm-ascend-workspace/vllm-ascend-pd-serving.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 99 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 639 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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00099 $0.00639
Opus 5 $0.00049 $0.00319
Sonnet 5 $0.00020 $0.00128
Haiku 4.5 $0.00010 $0.00064

Measured 11d ago against content hash cc105c29c328, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

vllm-ascend-pd-serving 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 11d ago.

The scan reads SKILL.md. This mod also ships 2 executable files (scripts/pd_serving.py, tests/test_pd_serving.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/vllm-ascend-pd-serving/SKILL.md · 63 lines

How it starts

The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.

vLLM Ascend PD Serving

Operate one prefill/decode deployment across an existing Session Group whose members have distinct session IDs. Reuse single-node Serving for each vLLM process; this Skill owns only cross-service role configuration, ordering, proxy health, rollback, and end-to-end smoke.

Preconditions

  • Every service has its own ready session.
  • session-management has grouped those sessions and proved an identical code plus submodule snapshot.
  • Connector type and options are explicit in the config.
  • The proxy or load balancer already has a stable URL. Proxy process lifecycle remains outside this MVP; its health and request path are verified here.

Workflow

  1. Run scripts/pd_serving.py plan with a PD config and Session Group file.
  2. Review generated commands. Connector options must already be present in each role's vLLM arguments; the controller never invents connector metadata.
  3. Run start. Services launch in declared order through the existing single-node Serving entry point.
  4. If any role fails, the controller stops already-started roles in reverse order and returns a failed state.
  5. Run status to inspect every member and the proxy health endpoint.
  6. Run smoke to send the configured request through the proxy and preserve the response as KV-transfer path evidence.
  7. Run stop; roles stop in reverse startup order.

Entry point

scripts/pd_serving.py provides plan, start, status, smoke, and stop.

Read only the reference needed for the active phase:

Boundaries

  • Single-node start/status/stop belongs to vllm-ascend-serving.
  • Session creation, leases, and group teardown belong to session-management.
  • A working PD deployment's accuracy or performance belongs to the corresponding validation workflow.
  • Hangs, rank divergence, endpoint mismatch, or connector diagnosis after a stable reproduction belongs to vllm-ascend-distributed-debug.

Read the full file on GitHub · 63 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.

  1. 11d ago First seen · 63 lines · 99 tokens per session scan A cc105c29c328

Subscribe to this mod's changes

vllm-ascend-pd-serving is a skill published in the GitHub repository maoxx241/vllm-ascend-workspace (36 stars, last pushed 7d ago), licensed MIT. It adds 99 tokens to every session and 639 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.

Related

Other skills, from other repositories

agent-platform-rag-engine-management

Manage and query Agent Platform RAG Engine Corpora and retrieve grounded contexts using the Google GenAI SDK. Use when listing RAG corpora or files, inspecting a corpus, retrieving contexts, or generating content grounded in a RAG corpus. Do not use for standard database queries (use SQL/Spanner skills), Google…

google/skills · 85 tokens

agent-platform-model-registry

Agent Platform Model Registry Management. Use when you need to upload, list, describe, update, or delete machine learning models (and their versions) in the Agent Platform Model Registry. Don't use for model training, model deployment to endpoints, or managing non-Agent Platform models.

google/skills · 60 tokens

foundry-config-setup

Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.

microsoft/agent-framework · 65 tokens

google-cloud-solution-agentic-analytics-spark-knowledge-catalog

Discovers requirements and generates guidance to design and deploy a governed, secure agentic-analytics solution for data that's distributed across Google Cloud, other cloud providers, or on-premises. Data that's outside Google Cloud (such as data from Databricks, Snowflake, Salesforce, SAP, or Oracle systems) is…

google/skills · 138 tokens

training-check

Interactively monitor training metrics from the current Codex session, periodically checking WandB or fallback logs for NaN, divergence, plateaus, and broken runs.

wanshuiyin/Auto-claude-code-research-in-sleep · 35 tokens

nemo-automodel-launcher-config

Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.

NVIDIA/skills · 30 tokens