fanar-mcp
Server Configuration
Describes the environment variables required to run the server.
| Name | Required | Description | Default |
|---|---|---|---|
| FANAR_API_KEY | Yes | Your Fanar API key. Required to use the server. | |
| FANAR_BASE_URL | No | Base URL for the Fanar API. | https://api.fanar.qa/v1 |
| FANAR_MCP_OUTPUT_DIR | No | Directory where audio and images are saved. | ~/fanar-mcp-output |
Instructions
Guidance the server publishes about itself, which clients place ahead of the tool catalog so the model reads it before choosing anything.
This server publishes no instructions, or was last inspected before Glama recorded them.
Capabilities
Features and capabilities supported by this server
Protocol revision2025-11-25
| Capability | Details |
|---|---|
| tools | {
"listChanged": false
} |
| prompts | {
"listChanged": false
} |
| resources | {
"subscribe": false,
"listChanged": false
} |
| experimental | {} |
Tools
Functions exposed to the LLM to take actions
| Name | Description |
|---|---|
| fanar_list_modelsA | List the Fanar models this API key may call. Entitlements differ per key: for example Fanar-Sadiq-2, which powers the quotation validator and deep research, must be enabled by Fanar. |
| fanar_chatB | Chat with a Fanar language model. Fanar-C-2-27B is the Fanar 2.0 27B model for Arabic (MSA and Gulf, Levantine and Egyptian dialects) and English; "Fanar" is the default alias. Set thinking=true for step-by-step reasoning. About 50 requests a minute. |
| fanar_islamic_qaA | Ask Fanar-Sadiq an Islamic question and get an answer grounded in cited sources such as the Qur'an, hadith collections and fatwa collections. Show the references with the answer and present it as a cited answer, never as a fatwa. Fanar-Sadiq-2 adds madhab and source filters when your key is entitled to it. About 50 requests a minute. |
| fanar_validate_quotationsA | Detect and verify Qur'an and hadith quotations inside any text: an LLM answer, a draft, an article. Verified quotations come back tagged and linked to canonical sources; a quotation left untagged could not be confirmed and must be treated as unverified. Requires Fanar-Sadiq-2 on the key. About 200 requests a minute. |
| fanar_deep_researchA | Start a Fanar-Sadiq-2 deep research job: a structured report on an Islamic topic, with headings and cited sources. It takes minutes (quick about 3-6, standard about 7-10); poll it with fanar_deep_research_result. Requires Fanar-Sadiq-2 on the key. About 20 jobs a day. |
| fanar_deep_research_resultB | Fetch the status, progress or finished report of a Fanar-Sadiq-2 deep research job. |
| fanar_translateA | Translate between Arabic and English with Fanar's Shaheen models, including classical Arabic such as hadith and fiqh texts. About 20 requests a day; keep each request under about 4,000 words. |
| fanar_text_to_speechA | Speak Arabic or English text with a Fanar Aura voice and save it as an audio file; returns the file path. Qur'an recitation by a named reciter needs Fanar-Sadiq-TTS-1. About 20 requests a day. |
| fanar_transcribeA | Transcribe Arabic (including dialects) or English audio from a local file with Fanar Aura. long_form=true uses the long-audio model (about 10 requests a day); otherwise about 20 a day. |
| fanar_generate_imageA | Generate a culturally aware image with Fanar Oryx and save it; returns the file path and the prompt as Fanar revised it. About 20 requests a day. |
| fanar_moderateA | Score a prompt and response pair with FanarGuard for safety and cultural awareness, in Arabic or English. Requires Fanar-Guard-2 on the key. |
| fanar_poemC | Compose Arabic poetry in the classical tradition with Fanar-Diwan. Requires Fanar-Diwan on the key. |
Prompts
Interactive templates invoked by user choice
| Name | Description |
|---|---|
| islamic_finance_question | Answer an Islamic finance question with cited sources and verified quotations. |
| verify_islamic_quotations | Check every Qur'an and hadith quotation in a text before it is published. |
Resources
Contextual data attached and managed by the client
| Name | Description |
|---|---|
No resources | |
TDQS
Scored across 12 tools
Each tool targets a distinct modality or task (chat, QA, translation, TTS, STT, image, moderation, poetry), and the deep_research/deep_research_result pair is clearly start-vs-poll. The only mild overlap is fanar_chat vs fanar_islamic_qa, since both answer questions with a language model, though the descriptions do distinguish grounded-citation QA from general chat.
All tools use a consistent fanar_ snake_case prefix with a predictable verb/noun structure (list_models, validate_quotations, generate_image, deep_research_result). A few names are bare verbs or nouns (chat, translate, transcribe, moderate, poem) rather than verb_noun, but the convention is still readable and uniform.
Twelve tools is well-scoped for a multimodal AI API, with one tool per capability and a clean async pair for deep research. No redundant or filler tools appear to pad the count.
The surface covers the major modalities (chat, QA, quotation validation, research, image, translation, TTS, STT, moderation, poetry) plus model listing and a full start/poll lifecycle for deep research. Minor gaps exist, such as no job cancellation/listing or embedding endpoint, but core workflows are covered.