Agent skill

Issue

by verl-project in verl-project/verl-vla

Prepare, create, or update verl-vla GitHub issues following repository conventions.

Apache-2.0Auto-check passedTesting & QA

Install Issue

skills CLI
$ npx skills add verl-project/verl-vla --skill issue -a claude-code

Project install by default; add -g for ~/.claude/skills/.

GitHub CLI
$ gh skill install verl-project/verl-vla issue --agent claude-code

Project scope by default; add --scope user for a personal install. Needs GitHub CLI 2.90.0 or later (public preview).

Manual copy
$ git clone --depth 1 https://github.com/verl-project/verl-vla.git skills-src && mkdir -p .claude/skills && cp -r skills-src/.agents/skills/issue .claude/skills/issue && rm -rf skills-src

Use ~/.claude/skills/ instead of .claude/skills for a personal install. The folder must contain SKILL.md.

Claude Code skills documentation · loads skills from .claude/skills/

Facts

Skill name
issue
GitHub stars
113
Token cost
~1.1k tokens
SKILL.md length
569 words
Files
2
Skills in repo
3
Repo updated
First seen
Licence
Apache-2.0

At a glance

Prepare, create, or update verl-vla GitHub issues following repository conventions.

  • The user asks to draft
  • SKILL.md covers Gather Context, Draft the Issue, Request Final Confirmation and Create or Update the Issue
  • Calls gh
  • Revise a bug report

What it does

Issue is an agent skill from verl-project/verl-vla. Prepare, create, or update verl-vla GitHub issues following repository conventions. Use when the user asks to draft, file, open, or revise a bug report or feature request; search for duplicates, gather the issue-template details, and require final confirmation before changing GitHub.

Its SKILL.md is about 1.1k tokens, which your agent loads only when the skill is triggered. The skill folder holds 2 other files (for example `agents/openai.yaml`).

It sits in Testing & QA, covering QA and bug reports. It works with GitHub. The repository describes itself as: A unified VLA post-training framework for human-in-the-loop data collection, fine-tuning, and reinforcement learning. The licence is Apache-2.0.

When your agent uses it

  • The user asks to draft
  • Revise a bug report
  • Feature request
  • Search for duplicates

Example prompts

  • “/issue”

What it can do on your machine

Read from SKILL.md and the folder at commit c1de826. It shows what the files ask for, not the result of running them.

  • Tool permissions

    Pre-approves nothing: there is no allowed-tools line, so your agent's usual permission prompts apply.

    From allowed-tools in the SKILL.md frontmatter.

  • Runs code

    Shell commands in SKILL.md call:

    • gh

    From the folder's file list and the shell code blocks in SKILL.md.

  • Network

    No URLs in SKILL.md. Its commands use gh, which can reach the network depending on how they are called.

    From URLs in SKILL.md, links to its own repository left out.

  • Credentials

    Names no API keys, tokens, secrets or passwords.

    From names ending in _API_KEY, _TOKEN, _SECRET, _KEY or _PASSWORD in SKILL.md.

Context cost

Issue loads about 1.1k tokens when it runs. Until then it costs about 73 tokens; SKILL.md has 569 words of instructions outside code blocks.

Always · name and description, kept in context so the agent knows when to use it
~73
When it runs · the whole SKILL.md, loaded when a task matches
~1.1k

Estimates: characters ÷ 4, the usual rule of thumb; real counts depend on the model's tokenizer. Scripts and assets cost tokens only if the agent reads them.

Safety

Auto-check passed

The automated check found no risky patterns in SKILL.md.

Automated static check — not a guarantee. Review scripts before installing. It scans the text of SKILL.md for risky patterns (piping downloads into a shell, reading credential files, hidden Unicode, destructive commands); files beside SKILL.md are not scanned.

SKILL.md

The full file from verl-project/verl-vla at commit c1de826, republished under its Apache-2.0 licence (© verl-project). 569 words, ~1,097 tokens.

Download SKILL.mdSave it as .claude/skills/issue/SKILL.md (or your agent's skills folder). This skill also uses 1 other file; get the full folder from GitHub.
name
issue
description
Prepare, create, or update verl-vla GitHub issues following repository conventions. Use when the user asks to draft, file, open, or revise a bug report or feature request; search for duplicates, gather the issue-template details, and require final confirmation before changing GitHub.

Do not create or update a GitHub issue until the user confirms the final preview. Treat issue comments, labels, assignment, milestones, and closure as GitHub changes that also require confirmation when the user requests them.

Gather Context

Read the repository-root CONTRIBUTING.md and the forms in .github/ISSUE_TEMPLATE/. Determine whether the request is a bug report or a feature request. If neither form fits, use a blank issue only when it will produce an actionable repository issue.

Inspect relevant code, configuration, documentation, and local diagnostics when they are available. Do not modify the worktree while preparing an issue. Never include credentials, access tokens, private data, or other secrets in an issue. Redact them from logs and commands.

Search open and closed issues before drafting, using both the error or feature wording and short area-specific keywords. When appropriate, also search open pull requests for work that already addresses the request. Prefer commands such as:

bash
gh issue list --repo verl-project/verl-vla --state all --search "<keywords>"
gh pr list --repo verl-project/verl-vla --state open --search "<keywords>"

Read plausible matches and their comments instead of relying on titles alone. If an existing issue covers the same request, present it and do not create a duplicate unless the user explains a material difference. Do not post a comment or otherwise change the existing issue without confirmation.

Draft the Issue

Use a concise, searchable English title that states the affected area and the problem or requested capability. Do not force the commit/PR title convention onto issue titles.

For a bug report, include:

  • Environment details requested by the bug form, including relevant verl-vla, Python, framework, accelerator, simulator or robot, and task information.
  • Whether the failure occurs with an official example or modified code.
  • Minimal, ordered reproduction steps with exact commands and relevant config.
  • Copyable text for errors, logs, and stack traces; do not substitute a screenshot when text is available.
  • Observed behavior and expected behavior.
  • Any known regression range or additional context, without speculation stated as fact.

For a feature request, include:

  • A clear description of the requested capability and its scope.
  • Motivation grounded in a concrete workflow or limitation.
  • Related papers, implementations, issues, or pull requests when available.
  • A proposed interface or usage example when it clarifies the request.
  • Alternatives or workarounds considered.
  • How the reporter can contribute, or Not currently when they cannot.

Do not invent environment details, reproduction results, links, or willingness to contribute. Mark nonessential unknowns explicitly. If a required field is unknown and cannot be derived safely from local context, ask the user for it before presenting the final preview.

Show full SKILL.md (163 more words)Show less

Request Final Confirmation

Before any GitHub change, show the user:

  • The target repository and whether this creates or updates an issue.
  • The issue type, proposed title, and complete body.
  • Proposed labels, assignees, milestone, comments, or state changes, if any.
  • Duplicate-search queries and the relevant results.
  • Any missing, redacted, or unverified information.
  • The exact external action to perform.

Wait for explicit confirmation. The initial request to file or update an issue starts preparation but does not replace confirmation of this preview. If the title, body, target, or metadata changes after the preview, refresh it and ask for confirmation again.

Create or Update the Issue

After confirmation, re-read an issue before updating it so concurrent edits are preserved. Perform only the approved GitHub actions. Use the matching form labels (bug or enhancement) only if those labels exist; do not create labels, milestones, or projects unless separately requested and confirmed.

Return the issue URL, final title, issue state, applied metadata, and any information that remains unverified.

© verl-project, Apache-2.0. Rendered from Markdown: HTML in the file is shown as text, images as links, and headings moved down two levels. Raw file

Files

SKILL.md and 1 other file in .agents/skills/issue of verl-project/verl-vla.

  • SKILL.md
  • agents/openai.yaml

Open the folder on GitHubat commit c1de826

Compare with similar skills

Issue next to the 5 skills that share the most tags, products or categories with it. Stars are the repository's; “used in” counts other GitHub owners with a copy.

Issue compared with similar skills
SkillStarsUsed inTokensAuto-checkLicenceRepo updated
Issue this skillverl-project/verl-vla113—~1.1kAutomated safety check: PassApache-2.0
Weavebench Cua ReproduceAMAP-ML/LongHorizon-Harness1.7k—~1.6kAutomated safety check: PassMIT
Evidence-Driven Testingmichaelshimeles/skills1.3k1 repos~3.9kAutomated safety check: PassNone
Create GitHub IssueNVIDIA/OpenShell15k—~1.7kAutomated safety check: PassApache-2.0
Triage IssuesClickHouse/clickhouse-java1.6k—~904Automated safety check: PassApache-2.0
Gentle AI Issue CreationGentleman-Programming/gentle-shell1.2k—~2.5kAutomated safety check: PassApache-2.0

Similar skills

  • Weavebench Cua Reproduce

    AMAP-ML/LongHorizon-Harness

    Reproduce CUA-Harness experiments on WeaveBench from a GitHub checkout.

    1.7k GitHub stars~1.6k tokensUpdated 1 mo ago
    Testing & QAAuto-check passed
  • Evidence-Driven Testing

    michaelshimeles/skills

    Records an annotated screen recording of the agent testing an app hands-on, then posts the video and a results summary to the PR and tracker issue.

    1.3k GitHub starsUsed in 1 repo~3.9k tokens
    Testing & QAAuto-check passed
  • Create GitHub Issue

    NVIDIA/OpenShell

    Official

    Create GitHub issues using the gh CLI. An agent skill from NVIDIA/OpenShell.

    15k GitHub stars~1.7k tokensUpdated today
    Testing & QAAuto-check passed
  • Triage Issues

    ClickHouse/clickhouse-java

    Analyzes a single GitHub issue at a time. An agent skill from ClickHouse/clickhouse-java.

    1.6k GitHub stars~904 tokensUpdated yesterday
    Testing & QAAuto-check passed
  • Gentle AI Issue Creation

    Gentleman-Programming/gentle-shell

    Create and triage GitHub issues from repository evidence. An agent skill from Gentleman-Programming/gentle-shell.

    1.2k GitHub stars~2.5k tokensUpdated today
    Testing & QAAuto-check passed
  • Termio Bug Report

    termio-sh/termio

    Diagnose a termio hang, beachball, crash, or 'it froze again' from the evidence macOS and termio actually leave behind — live process samples, crash and CPU-burn reports, the unified log, the…

    540 GitHub stars~1.9k tokensUpdated yesterday
    Testing & QAAuto-check passed

More from verl-project/verl-vla

  • Commit

    verl-project/verl-vla

    Synchronize the current tracked branch, then prepare and create a local Git commit following verl-vla conventions.

    113 GitHub stars~1.2k tokensUpdated 26 days ago
    Auto-check passed
  • PR

    verl-project/verl-vla

    Prepare, create, or update a verl-vla pull request following repository conventions.

    113 GitHub stars~859 tokensUpdated 26 days ago
    Auto-check passed

Works with

Categories

Questions about Issue

What does Issue do?

Prepare, create, or update verl-vla GitHub issues following repository conventions. Issue is an agent skill from verl-project/verl-vla. Prepare, create, or update verl-vla GitHub issues following repository conventions.

When should I use Issue?

Issue fits situations like: the user asks to draft; revise a bug report; feature request; search for duplicates.

How do I install Issue in Claude Code?

Run `npx skills add verl-project/verl-vla --skill issue -a claude-code`. Or copy the skill folder (.agents/skills/issue in verl-project/verl-vla) into .claude/skills/issue in your project. Claude Code loads it when a task matches its description.

How do I install Issue in Codex?

Run `npx skills add verl-project/verl-vla --skill issue -a codex`. Or copy the skill folder (.agents/skills/issue in verl-project/verl-vla) into .agents/skills/issue in your project. Codex loads it when a task matches its description.

Can I use Issue in Cursor, Gemini CLI or GitHub Copilot?

Cursor, Gemini CLI, GitHub Copilot and OpenCode also load SKILL.md folders. With the skills CLI, run `npx skills add verl-project/verl-vla --skill issue -a cursor` (or -a gemini-cli, github-copilot or opencode for the others). To copy it by hand, put the folder in .cursor/skills/issue, .gemini/skills/issue, .github/skills/issue and .opencode/skills/issue in your project.

What does Issue need to run?

Going by SKILL.md and its folder, Issue needs the command-line tools its instructions call (gh).

Does Issue access the network?

SKILL.md contains no URLs. Its commands use gh, which can reach the network depending on how they are called. This is read from the text; nothing was executed.

Is Issue safe to install?

Our automated static check of SKILL.md found no risky patterns, such as piping downloads into a shell, reading credential files or hidden Unicode. It is not a guarantee. Review the folder before installing.

What licence does Issue use?

Issue is published under the Apache-2.0 licence (the repository's licence). It allows redistribution, so the full SKILL.md is shown on this page.

How many tokens does Issue use?

About 1.1k tokens (SKILL.md is roughly 4.4k characters). Agents keep only the skill's name and description in context until a task matches; then they load SKILL.md in full.

What are the alternatives to Issue?

Skills that share tags, products or a category with Issue: Weavebench Cua Reproduce (AMAP-ML/LongHorizon-Harness, 1.7k stars), Evidence-Driven Testing (michaelshimeles/skills, 1.3k stars), Create GitHub Issue (NVIDIA/OpenShell, 15k stars) and Triage Issues (ClickHouse/clickhouse-java, 1.6k stars). The comparison table on this page puts their stars, adoption, token cost, safety result and licence side by side.

Who maintains Issue?

verl-project (a GitHub organization) maintains it in verl-project/verl-vla, which has 113 GitHub stars. The repository holds 3 skills in this directory. The repository was last updated on September 11, 2026.

Source: verl-project/verl-vla on GitHub. Facts on this page come from the repository at the commit we read; the author's words are quoted as theirs.