# twitter/communitynotes

Documentation and source code powering Twitter's Community Notes

Repository: https://github.com/twitter/communitynotes
Canonical: https://ross.abutalabs.com/products/communitynotes
Homepage: https://twitter.github.io/communitynotes
Language: Python
License: Apache-2.0
License Family: permissive
Topics: crowdsourcing, twitter
Last push: 2026-08-20T18:51:47+00:00

## Health v2 (maintenance only)
Score: 76/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 98, release rhythm 35, longevity 100
- inputs: {"age_days": 2046, "days_push": 13, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1880, forks 347 (observed 2026-08-28T04:05:48.583735+00:00)

## What it is
The public repository for Twitter/X's Community Notes, hosting the open-source scoring algorithm code, documentation, and a template AI note writer. It transparently shares the algorithms that add helpful context notes to potentially misleading posts.

## Use cases
- study how community notes scoring algorithm works
- build an AI note writer using the community notes API
- analyze scoring model output for research
- understand how twitter crowdsources fact-checking
- contribute alternate scoring algorithm ideas
- read documentation about the community notes program

## When to choose
- you want to research or reproduce the Community Notes scoring algorithm
- you're building an automated note writer on the Note Writer API
- you need transparent documentation of how X handles misinformation

## When to avoid
- you want to modify the core scoring algorithm for production use, since APIs must stay stable and core code is developed internally
- you need a general-purpose misinformation detection tool out of the box
- you expect to merge minor changes to scoring code, which is rarely accepted

## Facets
- artifact type: application
- maturity: active
- function: machine-learning, nlp, analytics, documentation, llm-inference
- domain: social-media, artificial-intelligence
- platform: python, self-hosted
- tags: misinformation, crowdsourcing, scoring-algorithm, twitter, note-writer-api, transparency, web-server

## Member repositories
- twitter/communitynotes (main) score 76

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:48.583735+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T03:14:01.846885+00:00, confidence not recorded.
  - readme: https://github.com/twitter/communitynotes (fetched 2026-08-28T04:05:48.583735+00:00, sha 68be8654576d)
- Data as of 2026-08-30T08:39:29.467469+00:00.