Skip to main content
Glama
ccalderong

conapesca-db-mcp

by ccalderong
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Hybrid

conapesca-db-mcp

MCP server for querying the CONAPESCA historical landings database (2001–2026), covering both the Pacific and Gulf coasts of Mexico.

Exposes tools for extraction and consultation of standardized fishing landings data (avisos de arribo).

Tools available

Data access (raw extraction)

Tool

Description

get_estados(year?)

States with landings, optional year filter

get_species(year?, estado?, tipo_aviso?, top_n?)

Species list with total weight and value

species_count()

Unique scientific names classified by taxonomic level

get_landings(year?, estado?, especie?, tipo_aviso?, oficina?, limit?, group_by?)

Landing records or aggregates; group_by: "year", "estado", "litoral"

get_offices(estado?)

CONAPESCA offices with record counts

get_taxonomy(especie)

Taxonomy + FishBase traits for a species

Reporting (summaries)

Tool

Description

landings_by_year(estado?, tipo_aviso?)

Annual totals: weight, value, species count

landings_by_estado(year?, tipo_aviso?)

Totals by state

landings_by_fleet_type(year?)

MAYORES vs MENORES vs COSECHA

Core

Tool

Description

health_check()

DB connectivity check

schema_snapshot()

Column names, types, row count

Related MCP server: @globalfishingwatch/gfw-mcp-js

Quick start (development)

1. Install

pip install -e ".[dev]"

2. Load dev data (SQLite)

python scripts/load_dev_data.py \
  --csv /path/to/conapesca_landings_2001_2026.csv \
  --db dev/conapesca_dev.sqlite \
  --rows 200000   # optional sample for fast testing

3. Configure

cp .env.example .env
# .env already has USE_SQLITE=true by default

4. Run

python -m mcp_server

Production (MySQL)

Set in .env:

USE_SQLITE=false
CONAPESCA_DB_HOST=...
CONAPESCA_DB_USER=...
CONAPESCA_DB_PASSWORD=...
CONAPESCA_DB_NAME=...

The table expected in MySQL is conapesca_landings_historical with the same schema as the pipeline output.

Resources

  • conapesca://data-dictionary — full column descriptions

  • conapesca://coverage — temporal and geographic coverage notes

This server cannot be deployed

One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Maintenance

ActivityStale
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers

  • grant-mcp

    vertical-mcp
    A
    license
    B
    quality
    D
    maintenance
    MCP server for searching research grants across NSF (US), ERC (EU), and KRF/NRF (Korea) via a unified interface. NIH excluded—covered by existing connectors.
    3
    7 npm
    MIT
  • @globalfishingwatch/gfw-mcp-js by GlobalFishingWatch

    @globalfishingwatch/gfw-mcp-jsofficial

    GlobalFishingWatch
    A
    license
    Not graded
    quality
    C
    maintenance
    Access Global Fishing Watch data from any MCP-compatible AI assistant or terminal: search vessels, retrieve fishing activity and port events, look up marine regions, and compute aggregate statistics.
    12 npm
    1
    Apache 2.0

  • emovi-mcp

    Lalitronico
    A
    license
    A
    quality
    C
    maintenance
    MCP server for the ESRU-EMOVI 2023 social mobility survey in Mexico. Enables AI assistants to query weighted statistics, transition matrices, and explore variables from the survey using natural language.
    11
    MIT