API Key Guide¶
This page explains which provider credentials ComProScanner can use, what each one is used for, and how to generate or obtain them.
Overview¶
ComProScanner can work with three groups of external providers:
Which credentials do you actually need?
| Provider group | Requirement level |
|---|---|
| Publisher/content providers for article access | Optional for manual or local workflows, but required for automated article retrieval. |
| LLM providers for extraction, vision models and RAG | At least one required for extraction, vision models and RAG workflows. However, default models are different for extraction/RAG and vision-language models. |
| Default embedding provider for vector database creation | Required unless you configure a custom embedding provider. |
Use only the providers relevant to your workflow. You do not need every key.
Publisher Providers¶
Elsevier / Scopus¶
Environment variable: SCOPUS_API_KEY
Used for:
- Scopus-based metadata retrieval
- Elsevier article retrieval in XML format
How to get it:
- Create or sign in to your Elsevier developer account.
- Open the API key management area.
- Create a key for Scopus or content APIs.
- Copy the generated key into your
.envfile asSCOPUS_API_KEY.
ScienceDirect Institutional Token¶
Environment variable: SCIENCEDIRECT_INSTTOKEN
Optional. Only required when accessing subscription-based Elsevier full-text articles and figures from outside the university campus network (i.e., for remote/off-campus access).
Used for:
- Remote off-campus access to Elsevier full-text article retrieval
- Downloading subscription-gated figures from ScienceDirect when working outside the campus network
How to get it:
- Contact your institution's library or IT/research computing team and ask for the ScienceDirect Institutional Token (insttoken).
- Elsevier issues this token per institution; individual researchers cannot generate it directly.
- Once obtained, copy it into your
.envfile asSCIENCEDIRECT_INSTTOKEN.
Springer Nature Open Access API¶
Environment variable: SPRINGER_OPENACCESS_API_KEY
Used for:
- Springer Open Access article retrieval in XML format
How to get it:
- Create or sign in to your Springer Nature account.
- Fill up the form to request an Open Access API key at https://dev.springernature.com/register/.
- Get the Open Access API key from the Springer Nature API management page.
- Copy the key into your
.envfile.
Springer Nature TDM API¶
Environment variable: SPRINGER_TDM_API_KEY
Used for:
- Springer subscription article retrieval in XML format
How to get it:
- Subscribe to the Springer Nature TDM service via https://dev.springernature.com/subscription/ and select the appropriate access level based on your institution and use case.
- Copy the issued TDM key or token into your
.envfile.
Wiley TDM API¶
Environment variable: WILEY_API_KEY
Used for:
- Wiley full-text article download as PDF
How to get it:
- Create your Wiley account.
- Login to your Wiley account at https://onlinelibrary.wiley.com/library-info/resources/text-and-datamining under the "Get a Text and Data Mining Token" section.
- Accept the terms and conditions to generate your API token.
- Copy the API token into your
.envfile.
IOP Publishing¶
Environment variable: IOP_papers_path (not an API key but a required path variable for processing IOP Science XML files)
Used for:
- Local processing of IOP Science XML files downloaded in bulk
How to get it:
- Email contentsupport@ioppublishing.org to request bulk access to the IOP Science XML files, typically through SFTP as IOP Publishing does not provide direct API access for bulk downloads.
- Once you have access, download the XML files to a local directory.
- Set
IOP_papers_pathto the absolute local folder path containing all the downloaded files.
LLM Providers¶
These providers can be used for extraction models, RAG chat models, and vision-language models where supported by your configuration.
OpenAI¶
Environment variable: OPENAI_API_KEY
Typical model prefixes: openai/... or OpenAI model names directly
How to get it:
- Create or sign in to your OpenAI account.
- Open the API keys section.
- Create a new secret key.
- Store it in
.env.
Google Gemini¶
Environment variable: GEMINI_API_KEY
Typical model prefixes: gemini/...
How to get it:
- Create or sign in to your Google AI Studio account.
- Generate an API key from the Gemini API key page.
- Store it in
.envasGEMINI_API_KEY.
Anthropic¶
Environment variable: ANTHROPIC_API_KEY
Typical model prefixes: anthropic/...
How to get it:
- Create or sign in to your Anthropic Console account.
- Create a new API key from the Anthropic keys page.
- Store it in
.env.
DeepSeek¶
Environment variable: DEEPSEEK_API_KEY
Typical model prefixes: deepseek/...
How to get it:
- Create or sign in to your DeepSeek platform account.
- Generate an API key from the DeepSeek API keys page.
- Store it in
.env.
OpenRouter¶
Environment variable: OPENROUTER_API_KEY
Typical model prefixes: openrouter/...
How to get it:
- Create or sign in to your OpenRouter account.
- Generate an API key from the OpenRouter keys page.
- Store it in
.env.
Together AI¶
Environment variable: TOGETHER_API_KEY
Typical model prefixes: together/...
How to get it:
- Create or sign in to your Together AI account.
- Generate an API key from the Together AI API keys page.
- Store it in
.env.
Cohere¶
Environment variable: COHERE_API_KEY
Typical model prefixes: cohere/...
How to get it:
- Create or sign in to your Cohere account.
- Create an API key from the Cohere API keys page.
- Store it in
.env.
Fireworks AI¶
Environment variable: FIREWORKS_API_KEY
Typical model prefixes: fireworks/...
How to get it:
- Create or sign in to your Fireworks AI account.
- Generate an API key from the Fireworks AI API keys page.
- Store it in
.env.
Ollama¶
Environment variable: none required
Used for:
- Local model inference
How to set it up:
- Install Ollama from the main Ollama website.
- Pull the model you want to use by following the Ollama library and setup docs.
- Set
base_urlorrag_base_urlif needed, such ashttp://localhost:11434.
Default Embedding Provider¶
Hugging Face¶
Environment variable: HF_TOKEN
Optional. Only required for downloading gated or private Hugging Face models. Public models work without a token.
Used for:
- Accessing gated or private Hugging Face models
- Rate-limited API access
How to get it:
- Create or sign in to your Hugging Face account.
- Open the access tokens page.
- Create a new token with the required permissions.
- Store it in
.env.
Recommended .env Template¶
Use the subset you need:
# Publisher providers
SCOPUS_API_KEY=your_scopus_api_key
SCIENCEDIRECT_INSTTOKEN=your_sciencedirect_insttoken # optional, for off-campus access
SPRINGER_OPENACCESS_API_KEY=your_springer_openaccess_api_key
SPRINGER_TDM_API_KEY=your_springer_tdm_api_key
WILEY_API_KEY=your_wiley_api_key
IOP_papers_path=/absolute/path/to/iop_papers
# LLM providers
OPENAI_API_KEY=your_openai_api_key
GEMINI_API_KEY=your_gemini_api_key
ANTHROPIC_API_KEY=your_anthropic_api_key
DEEPSEEK_API_KEY=your_deepseek_api_key
OPENROUTER_API_KEY=your_openrouter_api_key
TOGETHER_API_KEY=your_together_api_key
COHERE_API_KEY=your_cohere_api_key
FIREWORKS_API_KEY=your_fireworks_api_key
# Model and embedding access
HF_TOKEN=your_huggingface_token
Notes¶
- Keep all keys in your local
.envfile and never commit them to version control. - For most users, the minimum setup is one publisher source plus one LLM provider.
- If you use Gemini models, use
GEMINI_API_KEY. - If you use the default embedding setup, make sure
HF_TOKENis available.