Skip to main content
Glama
sandraschi

glama-status-mcp

by sandraschi
OverviewSchemaRelated ServersScoreDiscussions
TypeScript
Local

glama-status-mcp

FastMCP Tests Ruff TypeScript License Tauri

Daily-refreshed Glama TDQS score tracker for any MCP fleet. Scrapes per-tool grades from glama.ai, stores in SQLite with snapshot history and delta tracking, surfaces via MCP tools, Prefab cards, a 6-page web dashboard, and an LLM-powered chat interface. Tracks repos from any GitHub/Glama author via configurable config/fleet-repos.json.

Quickstart

git clone https://github.com/sandraschi/glama-status-mcp.git
cd glama-status-mcp

just install
.\start.ps1          # Full stack + browser at :11073
just refresh         # Manual scrape + snapshot

Related MCP server: mcp-federated-data

MCP Tools

Tool

Type

Description

glama_status

Portmanteau

11 ops: list, get, worst_tools, refresh, history, staleness, report, deltas, add_repo, remove_repo, reload_config

glama_scores_summary

Read-only

Compact grade distribution + per-repo stats

glama_daily_report

Read-only

Full markdown report: grades, deltas, worst tools, stale repos

glama_agentic_analyze

Mutating

Uses connected LLM (ctx.sample) to analyze scores and generate actionable fix todos

glama_generate_reports

Mutating

Writes per-repo markdown fix-todo reports to reports/

show_glama_status_card

Prefab card

Fleet overview dashboard card

show_glama_repo_card

Prefab card

Per-repo score breakdown card

Prompts

  • glama_improvement_plan(repo_name) -- Per-tool fix priorities with dimension scores

  • glama_fleet_analysis_prompt() -- Fleet-wide health snapshot for LLM ingestion

Web Dashboard (6 pages)

Page

Description

Dashboard

Hero section, sortable fleet table, click-to-drill repo detail with per-tool 6-dimension bars + Glama links

Report

Grade distribution bars, score deltas, worst tools fleet-wide, stale repo flagging

Tools

All tools grouped by grade (A-F), each linked to Glama tool page

Chat

LLM chat with 4 personalities, provider/model auto-discovery, conversation history, export

Help

5-tab documentation: Overview, Scoring, MCP Tools, REST API, FAQ

Settings

GitHub/Glama account config, persisted in localStorage

Scored Repos

10 demo repos with actual Glama scores. Add yours via glama_status(add_repo) or edit config/fleet-repos.json.

Repo

Grade

Score

Tools

blender-mcp

C

2.70

67

windows-operations-mcp

B

3.00

17

virtualization-mcp

B

3.06

9

worldlabs-mcp

B

3.38

20

robotics-mcp

A

3.58

8

bumi-mcp

A

3.64

2

xkcd-mcp

A

3.67

6

cursor-mcp

A

3.80

6

steam-mcp

A

3.81

14

email-mcp

A

3.82

10

Scoring (Glama TDQS)

Server score = 60% weighted mean + 40% minimum across tools. One low-scoring tool pulls the entire server grade down.

Grade

Threshold

A

>= 3.5

B

>= 3.0

C

>= 2.0

D

>= 1.0

F

< 1.0

Architecture

src/glama_status_mcp/
  server.py     -- FastMCP server + FastAPI HTTP + 7 tools + 2 prompts
  scraper.py    -- Async HTML scraper for glama.ai score pages
  database.py   -- SQLite storage + snapshot/delta engine (6 tables)
  models.py     -- Pydantic models + configurable fleet-repos.json loader
  llm.py        -- Glom-On auto-discovery (Ollama/LM Studio/OpenAI)
  config.py     -- Paths, constants
webapp/         -- Vite + React + TailwindCSS v4 (6 pages, 12 components)
native/         -- Tauri 2.0 wrapper + NSIS installer
reports/        -- Generated per-repo markdown fix-todo reports
config/         -- fleet-repos.json (user-editable)
tests/          -- 35 pytest tests + Playwright E2E

Install

Path

Audience

INSTALL.md

All installers: Option C (manual), Option D (dev), Options A-B (coming soon)

start.ps1

One-click: backend + frontend + browser

Native installer

Tauri NSIS: one download, one shortcut

.mcpb bundle

Drag into Claude Desktop

Ports

  • Backend (FastAPI + MCP HTTP /mcp): 11072

  • Frontend (Vite dev): 11073

Docs

File

Content

INSTALL.md

Full install guide

PRD.md

Product Requirements Document

CHANGELOG.md

Version history

llms.txt

LLM-facing index

llms-full.txt

LLM-facing full corpus

Install Server
A
license - permissive license
A
quality
B
maintenance

How are these scores calculated?

Maintenance

–Maintainers
–Response time
–Release cycle
–Releases (12mo)
Commit activity

Resources

Tools

Latest Blog Posts

MCP directory API

We provide all the information about MCP servers via our MCP API.

curl -X GET 'https://glama.ai/api/mcp/v1/servers/sandraschi/glama-status-mcp'

If you have feedback or need assistance with the MCP directory API, please join our Discord server