llamaIndex - 💡(How to fix) Fix [Feature Request]: list llama-index-readers-cvfile in LlamaHub catalog / docs

Feature Description

llama-index-readers-cvfile is already published on PyPI and implements BaseReader for the .cv open file format. The PyPI package is installable today (pip install llama-index-readers-cvfile) but is not surfaced anywhere on llamahub.ai or in the LlamaIndex docs, so users searching for resume or CV ingestion in a RAG pipeline cannot find it.

Requesting that the maintainers either:

1. Re index the catalog at https://llamahub.ai/l/readers/llama-index-readers-cvfile (currently returns 404 even though the package follows the documented [tool.llamahub] convention), or
2. Add an entry under the LlamaIndex docs example index pointing at the package, or
3. Clarify in the docs what the intended discoverability path is for community published readers so we can do the right thing on our side.

Package details:

• PyPI: https://pypi.org/project/llama-index-readers-cvfile/
• Source: https://github.com/cvfile/cv/tree/main/integrations/llama-index-readers-cvfile
• License: Apache-2.0
• Class: CVFileReader(BaseReader) returning list[Document] from a .cv file path
• pyproject.toml declares the canonical [tool.llamahub] metadata (import_path = "llama_index.readers.cvfile", contains_example = false, class_authors.CVFileReader = "cvfile.org")

Reason

The current CONTRIBUTING.md says new integration packages should live outside the monorepo and be published independently to PyPI. We did exactly that. What we cannot do from outside the repo is appear in the catalog, in the docs, or on llamahub.ai. A perfectly conformant reader is currently invisible to LlamaIndex users despite following every published rule.

Value of Feature

.cv is an open file format (PDF/A-3u carrying a Markdown payload plus pre computed BGE M3 vectors) designed for resume and CV ingestion. Recruiting, ATS, and hiring RAG pipelines built on LlamaIndex can use the reader to load a candidate resume and get a clean Document with the embedded Markdown as text. No OCR, no layout breakage, no re embedding. The format is open (CC BY 4.0 spec, Apache-2.0 SDKs), single purpose, and dependency light.

Full spec: https://cvfile.org/spec/ Home: https://cvfile.org