Find the Alita paper · VFS-Bench 1.0

Task Prompt

I'm looking for a paper on arxiv related to agentic reasoning. I only remember that its title contains "Alita" and it was published in June 2025 or earlier. Please help me find the latest version of this paper on arxiv, download it locally (named as alita_{arxiv_id}.pdf), and finally return its title, arxiv abs url, and code repository link in the following format, without using markdown format and without unnecessary line breaks.

title: {title}
arxiv_abs_url: {arxiv_abs_url}
code_url: {code_url}

name	workspace-filesystem
description	The virtual filesystem layout of this workspace.

Path	Description
`/workspace`	Agent workspace; current directory and home (`~`).
`/tmp`	Temporary files.
`/dev`	Device files: `null` and `zero` (mounted automatically).

SKILL.md

---
name: workspace-commands
description: The shell commands and task tools available in this workspace.
---

# Available commands

Python (`python` or `python3`), Perl, JavaScript/Node.js, and TypeScript
runtimes are unavailable. Use the shell commands and task CLIs documented
below; do not probe for or attempt to install other runtimes.

Standard Unix commands, among them:

    cat  head  tail  ls  find  grep  wc  sort  uniq  cut  tr  sed  awk
    diff  tee  xargs  jq  unzip

with pipes, `>` redirection, `$(...)` substitution and `$'...'`.

The task CLIs below are registered with this shell rather than installed
as files; do not inspect them with `which`, `file` or `/usr/bin`.

## arxiv-local

Use `arxiv-local <command> '<JSON arguments>'`. Run
`arxiv-local <command> --help` to get more information.

| Command | Arguments | Description |
|---|---|---|
| `arxiv-local search-papers` | `{query: str, max_results?: int, date_from?: str, date_to?: str, categories?: list[str]}` | Search for papers on arXiv with advanced filtering |
| `arxiv-local download-paper` | `{paper_id: str, check_status?: bool = false}` | Download a paper and create a resource for it — `paper_id`: The arXiv ID of the paper to download; `check_status`: If true, only check conversion status without downloading |
| `arxiv-local list-papers` | `{}` | List all existing papers available as resources |
| `arxiv-local read-paper` | `{paper_id: str}` | Read the full content of a stored paper in markdown format — `paper_id`: The arXiv ID of the paper to read |

## scholarly

Use `scholarly <command> '<JSON arguments>'`. Run
`scholarly <command> --help` to get more information.

| Command | Arguments | Description |
|---|---|---|
| `scholarly search-arxiv` | `{keyword: str}` | Search arxiv for articles related to the given keyword. |
| `scholarly search-google-scholar` | `{keyword: str}` | Search google scholar for articles related to the given keyword. |

name	workspace-commands
description	The shell commands and task tools available in this workspace.

Available commands

Python (python or python3), Perl, JavaScript/Node.js, and TypeScript runtimes are unavailable. Use the shell commands and task CLIs documented below; do not probe for or attempt to install other runtimes.

Standard Unix commands, among them:

cat  head  tail  ls  find  grep  wc  sort  uniq  cut  tr  sed  awk
diff  tee  xargs  jq  unzip

with pipes, > redirection, $(...) substitution and $'...'.

The task CLIs below are registered with this shell rather than installed as files; do not inspect them with which, file or /usr/bin.

arxiv-local

Use arxiv-local <command> '<JSON arguments>'. Run arxiv-local <command> --help to get more information.

Command	Arguments	Description
`arxiv-local search-papers`	`{query: str, max_results?: int, date_from?: str, date_to?: str, categories?: list[str]}`	Search for papers on arXiv with advanced filtering
`arxiv-local download-paper`	`{paper_id: str, check_status?: bool = false}`	Download a paper and create a resource for it — `paper_id`: The arXiv ID of the paper to download; `check_status`: If true, only check conversion status without downloading
`arxiv-local list-papers`	`{}`	List all existing papers available as resources
`arxiv-local read-paper`	`{paper_id: str}`	Read the full content of a stored paper in markdown format — `paper_id`: The arXiv ID of the paper to read

scholarly

Use scholarly <command> '<JSON arguments>'. Run scholarly <command> --help to get more information.

Command	Arguments	Description
`scholarly search-arxiv`	`{keyword: str}`	Search arxiv for articles related to the given keyword.
`scholarly search-google-scholar`	`{keyword: str}`	Search google scholar for articles related to the given keyword.

Model	VFS	MCP	VFS Tokens	MCP Tokens	Token Reduction Rate
Claude Opus 5.5	1/1	1/1	117k	169k	31%
GPT-6.1 Sol	1/1	1/1	15.2k	40.9k	63%
GPT-5.6 Luna	1/1	1/1	166k	75.3k	−120%
Claude Haiku 5.5	1/1	1/1	604k	414k	−46%
Gemini 3.8 Flash	1/1	1/1	799k	911k	12%

find-alita-paper

Task Prompt

VFS.md

SKILL.md