https://github.com/stemmlerjs/stemmlerjs-bot
Twitter bot growth hacking experiment to go from 300 to 3K followers
https://github.com/stemmlerjs/stemmlerjs-bot
bot experiment nodejs twitter
Last synced: 2 months ago
JSON representation
Twitter bot growth hacking experiment to go from 300 to 3K followers
- Host: GitHub
- URL: https://github.com/stemmlerjs/stemmlerjs-bot
- Owner: stemmlerjs
- Created: 2018-02-04T22:37:43.000Z (over 7 years ago)
- Default Branch: master
- Last Pushed: 2018-06-23T05:07:58.000Z (about 7 years ago)
- Last Synced: 2025-04-12T02:09:30.695Z (2 months ago)
- Topics: bot, experiment, nodejs, twitter
- Language: JavaScript
- Homepage: https://medium.com/@karanjthakkar/how-i-grew-from-300-to-5k-followers-in-just-3-weeks-2436528da845
- Size: 40 KB
- Stars: 0
- Watchers: 2
- Forks: 0
- Open Issues: 0
-
Metadata Files:
- Readme: README.md
Awesome Lists containing this project
README
# stemmlerjs-bot
## Layers
### Actuation Layer (Layer 0)
- ```/bot.js```
- This layer does all of the actions. It simply performs an action and logs the result of that action.
- Success or failure to do the action.
- It is not concerned with timing or anything.
### Scheduling Layer (Layer 1)
- ```scheduler.js```
- This layer is responsible for scheduling the actions to be performed.
- Actions to be performed come in through the decision layer.
- There is a max number of queued items that can possibly exist at one time (30).
- Actions can be cancelled, clearing that item from the queue and making room for a new action to be queued.
- This layer needs to be aware of the rate limits for events.
- It also needs to be pseudo-random with when it schedules the events to be actuated in the actuation layer.
### Decision Layer (Layer 2)
- ```decision.js```
- This layer decides whether or not the action should be performed or not.
- It looks into a collection of information to decide whether the action should be performed or not.
- Following / Follow ratio
- Interest keywords found in their bio or in their tweets.
- When it finds something that is particularly convincing such that we should follow this user, we will do a calculation and give points based on
what we find.
- It also filters all decisions that come in first. We don't want to follow any profiles that promote hateful content or porn.
### Search Layer (Layer 3)
- ```search.js```
- This layer actively searches for users to follow and tweets to like.
- There are a few different ways that we can begin to funnel content into this to search for.
- Seed Accounts: These are accounts that we know are highly related to the content that we are seeking.
- We will go through this user's list of followers and incrementally add them.
- Twitter Feed: We will watch for new tweets that match our interests and patch an action to the decision layer to decide whether or
not to follow that user.