vllm-ascend-workspace: Skill for Claude Code

.agents/skills/session-management/SKILL.md

session-management is a skill for Claude Code, Codex from maoxx241/vllm-ascend-workspace. It costs 62 tokens per session (2,982 once invoked), scanned A, original, MIT.

A manager for isolated VAWS agent sessions, where each session gets its own Git worktree, remote container, state directory, and resource leases. VAWS is the workspace system used to separate agent tasks and remote resources.

In plain words
What is it for?
Use it to create, list, inspect, group, remove, and clean up sessions, or coordinate several agents that need matching code and separate remote environments.
Why use it?
It prevents parallel tasks, services, containers, and hardware assignments from interfering with one another.

Skill for Claude CodeCodex

Written for no agent in particular: nothing here depends on one. Also seen: installed under .agents/ (shared by several agents).

This is maoxx241/vllm-ascend-workspace's own configuration. It tells Claude Code and 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/session-management/SKILL.md
Clone the repo
git clone --depth 1 https://github.com/maoxx241/vllm-ascend-workspace

Made for: Claude Code, 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 session-management

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

Your own site · 80×15
<a href="https://agentmods.dev/skills/maoxx241/vllm-ascend-workspace/session-management"><img src="https://agentmods.dev/badge/skills/maoxx241/vllm-ascend-workspace/session-management.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 62 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,982 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 $0.00062 $0.02982
Opus 5 $0.00031 $0.01491
Sonnet 5 $0.00012 $0.00596
Haiku 4.5 $0.00006 $0.00298

Measured 6d ago against content hash f8eb6572171f, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-10, from the pricing page.

Security

Grade A, and why

session-management 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 6d ago.

The scan reads SKILL.md. This mod also ships 16 executable files (scripts/npu_coordination.py, scripts/session_create.py, scripts/session_diff.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/session-management/SKILL.md · 193 lines

How it starts

The opening of the file, as written. The whole thing — 193 lines — stays where its author put it; the contents beside it link to each section on GitHub.

Session Management

Local task identity and native clients

For the task-facing mode, use remote-dev's vaws_session, vaws_run, vaws_execution and vaws_finish tools. The native adapter supplies context_file; the user should not have to manage it. Configure once using .agents/scripts/vaws_client_setup.py --client <client> --project <actual-repo> (preview, then --apply after reviewing the changes). This does not grant client trust or authentication. See lifecycle and client setup.

  • A new user-created native session creates a new VAWS task, even in the same cwd.
  • Native resume continues the same native attachment and VAWS task.
  • Spawned children inherit through a native parent id or explicit VAWS_PARENT_CONTEXT; a user-authorized new peer uses VAWS_ATTACH_CONTEXT. Never infer association from cwd, branch, chat text, or newest transcript.
  • Bind the actual repository worktrees with vaws_session(sources=...). Create repository worktrees with native Git tools only when code isolation is needed; never make another scaffold/submodule copy just to get a task id.
  • With no remote configuration, identity/source binding/native local tools remain available. vaws_run is the first operation that contacts the fleet.
  • vaws_run materializes a pinned source snapshot into an exclusive prepared runtime, then queues/activates a host lease and starts a supervised job. It does not create containers, install dependencies or compile on a cache miss.
  • Closing a native session only detaches it. Explicit task finish stops owned jobs, waits for process/device release, and preserves local worktrees.

The native hook and managed-job additions require separate regression and real-client acceptance; the earlier two-card pool probe is not their evidence.

Legacy container-bound sessions

When .vaws-local/workspace-identity.json contains a unified alias, new session container names inherit the base machine namespace. Session records also snapshot agent_id and alias for cooperative attribution. Existing session/container names are never rewritten after an alias change.

Read the full file on GitHub · 193 lines

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. 6d ago Changed · +66 lines · -35 tokens per session f8eb6572171f
  2. 10d ago First seen · 127 lines · 97 tokens per session scan A db5b21e620dc

Subscribe to this mod's changes

session-management is a skill published in the GitHub repository maoxx241/vllm-ascend-workspace (36 stars, last pushed 6d ago), licensed MIT. It adds 62 tokens to every session and 2,982 once invoked, about $0.0003 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.