---
name: paper-finder
description: Discover recent arXiv papers matching a research profile and generate Obsidian-compatible Markdown notes. Use this skill when the user wants to find new papers, refresh their literature inbox, or search arXiv based on their interests.
---
# Paper Finder

Use this skill when you want to discover recent arXiv papers for a specific research profile and write the results as Obsidian-compatible Markdown notes.

## Run Retrieval

Run the CLI from the repository root:

```bash
python skills/paper-finder/find_papers.py \
  --profile path/to/research-interest.json \
  --output path/to/obsidian/inbox
```

Optional semantic ranking:

```bash
python skills/paper-finder/find_papers.py \
  --profile path/to/research-interest.json \
  --output path/to/obsidian/inbox \
  --semantic
```

Semantic settings are read from `stable-jarvis/.env`:
- `EMBEDDING_BASE_URL`
- `EMBEDDING_API_KEY`
- `EMBEDDING_MODEL`

You can pass `--semantic-config path/to/config.json` for advanced overrides.

## Build or Refresh a Profile

If the user does not already have a profile JSON, use Zotero MCP read tools to collect evidence first:

1. `zotero_list_collections`
2. `zotero_profile_evidence`

Then write a JSON profile matching the same structure used by `config/research-interest.example.json`:

- `profile_id`
- `profile_name`
- `zotero_basis`
- `retrieval_defaults`
- `interests[]`

Prefer short `method_keywords` and only a small number of `query_aliases` per interest.

Write it in `temp/paper-finder/research-interest.json`, if there is no such directory, create it.

## Enrich Candidate Notes

After the Python retrieval run finishes, read the generated Obsidian note and use Zotero MCP read tools to gather nearby library evidence. Then update the note's frontmatter and the `Why It Matters`, `Quick Takeaways`, and `Caveats` sections using the prompt in `skills/paper-finder/prompts/enrich-candidate.prompt.txt`.

---
Source attribution: JaimeParker
Original project: https://github.com/JaimeParker/stable-jarvis/blob/cb2c7078e8b76a3125ae504aeb1ffa9c1ee2dc4b/skills/paper-finder/SKILL.md
License: MIT
License information: https://opensource.org/license/mit
Original upstream instructions; downloaded from Dave AI.

## Upstream license notice

MIT License

Copyright (c) 2026 Zhaohong Liu

Permission is hereby granted, free of charge, to any person obtaining a copy
of this software and associated documentation files (the "Software"), to deal
in the Software without restriction, including without limitation the rights
to use, copy, modify, merge, publish, distribute, sublicense, and/or sell
copies of the Software, and to permit persons to whom the Software is
furnished to do so, subject to the following conditions:

The above copyright notice and this permission notice shall be included in all
copies or substantial portions of the Software.

THE SOFTWARE IS PROVIDED "AS IS", WITHOUT WARRANTY OF ANY KIND, EXPRESS OR
IMPLIED, INCLUDING BUT NOT LIMITED TO THE WARRANTIES OF MERCHANTABILITY,
FITNESS FOR A PARTICULAR PURPOSE AND NONINFRINGEMENT. IN NO EVENT SHALL THE
AUTHORS OR COPYRIGHT HOLDERS BE LIABLE FOR ANY CLAIM, DAMAGES OR OTHER
LIABILITY, WHETHER IN AN ACTION OF CONTRACT, TORT OR OTHERWISE, ARISING FROM,
OUT OF OR IN CONNECTION WITH THE SOFTWARE OR THE USE OR OTHER DEALINGS IN THE
SOFTWARE.



# Dave AI resource metadata

Creator / publisher: JaimeParker
Original project: https://github.com/JaimeParker/stable-jarvis/blob/cb2c7078e8b76a3125ae504aeb1ffa9c1ee2dc4b/skills/paper-finder/SKILL.md
Attribution basis: Source repository record

Fields of work: Education & Research, Marketing
Roles: Academic Writer, Content Marketer, Researcher
Specialties: API, Backend, Word Documents, Frontend
Tools: Obsidian
Country applicability: Country not specified
Classification basis: Source catalog labels
Instruction SHA-256: cebc857deec2ab09056d2181d4785ebcc9ee398d2484333da524250a82ff84e8