SA
SkillAvatars
Back to MCP Tools

Archive Agent

data-ai MCP Server

Find your files with natural language and ask questions.

byshredEngineer
VerifiedReviewed
data-aidata-ai
3 views59 stars9 forksGPL-3.0
View on GitHubHomepageDocs

Why This Matters

Discovered via github-seeds:mcp-hot and last synced 3mo ago.

VerifiedReviewed
Source
github-seeds:mcp-hot
Stars
59
Last synced
3mo ago
Install
Check source

Install

Install instructions not detected yet

Check the source repository for the latest setup steps.

View source instructions
47
Tools
0
Resources
0
Prompts
Standard I/O
Transport

Available Tools (47)

Rerank

`google/gemini-2.5-flash-lite`

config_version

Config version

qdrant_collection

Name of the Qdrant collection

Query

`google/gemini-2.5-flash`

get_collections

None

Vision

`google/gemini-2.5-flash`

get_search_result

`search`

ai_model_rerank

AI model used for reranking

get_files_tracked

`track` and then `list`

get_answer_rag

`query`

chunk_lines_block

Number of lines per block for chunking

rerank_chunks_max

Number of top chunks to keep after reranking

ocr_auto_threshold

Minimum number of characters for `auto` OCR strategy to resolve to `relaxed` instead of `strict`

get_files_changed

`track` and then `diff`

Implementation

Description

get_chunk_headers

None

mcp_server_port

MCP server port (default `8008`)

retrieve_score_min

Minimum similarity score of retrieved chunks (`0`...`1`)

max_workers_embed

Maxmimum number of parallel embedding requests **per file**, creating one thread per request

Embed

`openai/text-embedding-3-large`

mcp_server_host

MCP server host (default `http://127.0.0.1`; set to `http://0.0.0.0` to expose in LAN)

max_workers_vision

Maxmimum number of parallel vision requests **per file**, creating one thread per request

Chunk

`google/gemini-2.5-flash-lite`

Synchronous

Get the list of available Qdrant collections.

Key

Description

image_entity_extract

Image handling: `true` enables entity extraction, `false` disables it.

chunk_words_target

Target number of words per chunk

ai_model_embed

AI model used for embedding

qdrant_server_url

URL of the Qdrant server

Task

Default Model

retrieve_knee_sensitivity

Knee detection sensitivity (Kneedle `S` parameter; higher = more conservative)

Asynchronous

Get answer to question using RAG.

ocr_strategy

OCR strategy in [`DecoderSettings.py`](archive_agent/config/DecoderSettings.py)

retrieve_chunks_max

Maximum number of retrieved chunks

ai_server_url

AI server URL

image_ocr

Image handling: `true` enables OCR, `false` disables it.

retrieve_knee_enable

Adaptive cutoff for retrieval (`true` enables knee-based cutoff, `false` disables it)

ai_provider

AI provider in [`ai_provider_registry.py`](archive_agent/ai_provider/ai_provider_registry.py)

ai_temperature_query

Temperature of the query model (ignored for GPT-5)

retrieve_knee_min_chunks

Minimum number of chunks to keep when adaptive cutoff is applied

ai_model_chunk

AI model used for chunking

ai_model_vision

AI model used for vision (`""` disables vision)

max_workers_ingest

Maximum number of files to process in parallel, creating one thread for each file

ai_model_query

AI model used for queries

ai_vector_size

Vector size of embeddings (used for Qdrant collection)

get_patterns

`patterns`

expand_chunks_radius

Number of preceding and following chunks to prepend and append to each reranked chunk

Browse all MCP servers