Career Routines logo

Career Routines

Community
Zal4DW
career-routines

Use when the user wants parts of their job search to run on a schedule, or asks about routines they already have, such as putting the weekly update on a Cowork schedule, setting up narrower recurring tasks, or working out why a scheduled task did not run. Explains when a task runs in the cloud and when it needs their computer.

Overview

PublisherZal4DW
Repositorycareer-helper
Skill namecareer-routines
Stars
116
Forks
15
Bundled files
3
Links
  • Markdown instructions

    A SKILL.md file the model loads on demand, so it only costs tokens when a request actually matches.

  • Works with any LLM

    AI skills are plain Markdown, not provider-specific code, so this works with GPT, Claude, Gemini, Grok, or a local model.

  • 3 bundled files

    Scripts, templates, and references the model can read while it works. Files are read-only and never executed.

  • Open source

    Published by Zal4DW on GitHub. Read the source before you install it.

Installation

Install the Career Routines AI skill in TypingMind to use it with any LLM, or drop it into another agent that reads SKILL.md.

1

Install in TypingMind

TypingMind installs a skill straight from its GitHub folder — it reads SKILL.md, bundles the resource files, and stores the result locally.

  1. Open the app and go to Plugins → Skills.
  2. Choose "Install from GitHub".
  3. Paste the skill folder URL below and confirm.
    https://github.com/Zal4DW/career-helper/tree/main/career-helper/skills/career-routines
  4. Enable the skill in any chat where you want it available.
Plugins → Skills → Add skill → From GitHub URL, then paste the folder URL and press Continue.
2

Install in another agent

Any agent that reads the Agent Skills format can use this skill — copy the folder into that agent's skills directory.

Claude Code — .claude/skills
git clone --depth 1 https://github.com/Zal4DW/career-helper.git /tmp/career-helper
mkdir -p .claude/skills
cp -r /tmp/career-helper/career-helper/skills/career-routines .claude/skills/career-routines
Restart Claude Code after copying so it picks up the new skill.

Use it in TypingMind

Enable Career Routines in any TypingMind chat and the model takes it from there. Its name and description sit in the system prompt, and the moment a request matches, the model loads the full instructions itself — you never invoke it by hand, and it costs no tokens until it is actually used.

The model loads Career Routines on its own as soon as a request matches it.

Works with any AI model

AI skills are plain Markdown instructions rather than provider-specific code, so Career Routines is not tied to the model it was written for. Install it once in TypingMind and use it with GPT-5, Claude, Gemini, Grok, DeepSeek, Mistral, Llama, or a local model you run yourself — all on your own API keys.

  • Loaded only when it is needed

    The system prompt carries just the name and description. The instructions are fetched on the first matching request, so an idle skill costs nothing.

  • Switch models mid-chat

    Because the skill is instructions rather than code, changing model does not break it — the next model reads the same SKILL.md.

Skill instructions

This is the SKILL.md content the model loads. Read it before installing — a skill is instructions your model will follow.

Career Routines

Put the recurring parts of your job search on a schedule, so the tracker, follow-ups, and market map keep themselves current between sessions.

Capabilities

#CapabilityWhen to Use
1Set Up a RoutineCreating a Claude Cowork scheduled task, tailored to your workspace and cadence
2Routine LibraryChoosing from the ready-made prompts, starting with the one-command weekly update
3Review Your RoutinesSeeing what is scheduled, what last ran, and fixing a routine that is not doing its job
4Outside CoworkScheduling the same prompts from Claude Code Desktop, the command line, or a cloud Routine

Quick Start

text
"Set up a weekly routine that keeps my job search moving"
"Schedule the weekly update for Monday mornings"
"Put my market map update on a timer"
"What routines do I have running?"
"My scheduled task did not run over the weekend. Why?"

Accessibility

At skill start, check for career-helper-preferences.md in the current working directory using the Glob tool. If found, read the YAML frontmatter and apply:

  • dyslexia_friendly: true: Use short sentences. Number all lists and options (never unnumbered). One decision per message. No idioms or metaphors; use plain replacements. Explicit signposting at every transition ("Step 2 of 3. Next: choosing the cadence."). Refer to saved files by description, not filename. Repeat key details (routine names, days, times); do not assume the user remembers from earlier messages.
  • colour_blind: true: Never use colour alone to convey meaning. Use labels, text, or icons for all status indicators.

If no preferences file exists and this skill was invoked directly (not dispatched by Tim): ask once, "Do you have any accessibility preferences I should know about? For example, if you're dyslexic, I can adjust how I format things." If yes, ask whether you may save them so they do not have to repeat them next time, and save to career-helper-preferences.md using the format in @../tim/references/tim-preferences-format.md only if they agree. If the user has no preferences or declines to have them saved, proceed without creating the file.

These rules apply to all communication with the user and to the formatting of output documents.


How Scheduling Works Here

Career Helper is designed for Claude Cowork on Claude Desktop, and Cowork has scheduled tasks built in. You write a prompt once, choose how often it runs, and Cowork runs it as its own session with the plugin's skills available. The plugin supplies the prompts and the one command that does all the updating; Cowork supplies the scheduler.

The one thing to understand before scheduling anything is where the task runs, because it decides whether your computer needs to be on:

  • In the cloud when the task does not need a folder on your computer. Your laptop can be closed. It can use connected tools, plugins, skills, web research, and files saved to your Claude account.
  • On your machine when the task is given a folder on your computer. Then your computer and Claude Desktop must be available at the scheduled time.

Career Helper works in both Desktop Cowork and cloud Cowork, but they keep data differently, and routines depend on that. Desktop Cowork with a local folder is the most feature-rich option for persistence: every session and every scheduled run sees the same files on disk. Cloud Cowork keeps created files in the conversation's data store, so a long-running piece of work there means staying in the same conversation, and a scheduled run (a new session each time) cannot build on last week's files unless the workspace lives on something every session can reach, such as a connected drive.

So the honest default for a routine that reads the tracker or market map is Desktop Cowork with the workspace folder, running on the user's machine at a time it is awake. Say this plainly before setting anything up, and let the user choose.

Load: @references/cowork-scheduling.md for the setup steps, the cloud-or-local rules, permissions, and troubleshooting.


1. Set Up a Routine

What you need: Your workspace folder, a Cowork session, and a rough idea of cadence Load: @references/cowork-scheduling.md Template: @references/routines-register-template.md

Walk the user through one routine at a time:

  1. Check what exists. Glob the workspace for applications/tracker.md, market-map.md, applications/learnings/, and any content calendar. A routine that reads a file which does not exist reports "not found" every week; suggest building the file first, or pick a routine whose inputs exist.
  2. Recommend the routine. For almost everyone the first routine is the weekly update (/career-helper:weekly-update), because it already contains the standup, the follow-up check, and the market map update. Offer the narrower prompts only for a different cadence.
  3. Tailor the prompt. Fill the bracketed parts (target role, location, employer where relevant). Keep the "do not invent" and "record decisions, do not ask" lines; they are what keep an unattended run honest.
  4. Decide where it runs. Ask whether they want it on their machine (Desktop Cowork, folder chosen, computer must be awake) or in the cloud (no local folder, lid closed, but no memory of last week's files unless the workspace is on a connected drive). State the trade-off in two sentences.
  5. Give the exact steps. Type /schedule in any Cowork task (or open Scheduled in the sidebar), paste the prompt, choose the cadence, choose the folder or leave it for the cloud, and save. Suggest running it once immediately to confirm it works and to approve anything it needs.
  6. Record it. Add a row to routines.md (create from the template if absent): name, cadence, where it runs, prompt, date set up. /career-helper:status reads this file.

Output: routines.md at the workspace root, plus the copy-paste prompt and setup steps in conversation


2. Routine Library

What you need: Nothing; the library is ready-made Load: @references/routine-library.md

Seven prompts, each safe to run unattended:

  1. Weekly update (the one command; start here)
  2. Monday job-search standup
  3. Weekly market and role monitor
  4. LinkedIn posting reminder
  5. Application follow-up check
  6. Pre-interview prep nudge (on demand)
  7. Weekly market map update (standalone, for those who only want the map)

Present the library only when asked, or when the weekly update does not fit; do not offer all seven to a new user.


3. Review Your Routines

What you need: routines.md, and updates/ or market-watch/ if the routines write there Load: @references/cowork-scheduling.md (Troubleshooting section)

Read routines.md and the output folders, then report:

  • Each routine, its cadence, where it runs, and the date of its most recent output file (or "no output found")
  • Anything that looks stale: a weekly routine with no output in two weeks
  • Open items from the latest weekly update's "Decisions waiting for you" section
  • The likely cause when a routine has not run, in order of frequency: the task was given a local folder and the computer was asleep; the task stalled on a permission it had not been granted; the input file it reads does not exist; the schedule was paused

Offer the fix for each, and update routines.md when something changes. Never guess at what a run did; read its output file or ask the user to open the run in Cowork's Scheduled list.

Output: Conversational review, updated routines.md


4. Outside Cowork

What you need: To know where the user runs Career Helper Load: @references/cowork-scheduling.md (Outside Cowork section)

Some people run the plugin in Claude Code rather than Cowork. The prompts are identical; only the scheduler changes:

  • Claude Code Desktop local scheduled task: Routines in the sidebar, New routine, Local; runs on the machine with the folder; needs the app open and the computer awake, with one catch-up run on wake
  • Command line with cron or launchd: claude -p "/career-helper:weekly-update" --permission-mode acceptEdits --permission-prompts none from the workspace folder; the machine must be awake; no catch-up
  • Cloud Routine (claude.ai/code/routines): runs without the machine but only clones a GitHub repository, so it suits a workspace kept in a private repository and nothing else

/loop in Claude Code repeats a prompt only while a session is open and is not a scheduler.


Coaching Voice

Routines are plumbing, and plumbing should be boring:

  • One routine first. A new user who schedules six things reads none of them. Start with the weekly update.
  • Be plain about where it runs. "Your laptop needs to be awake on Monday morning" is a sentence the user needs to hear before they rely on it.
  • A quiet report is a good report. If the weekly update says nothing changed, that is the routine working, not failing.
  • Do not let routines replace decisions. The unattended run records what it could not decide; the user still decides. Say so.

Output Standards

  • UK English throughout
  • No emojis; professional tone
  • Honest about limits: where a task runs, what it can reach, and what it cannot do unattended
  • Copy-paste ready: every prompt and every scheduler step is given verbatim
  • No invented state: routines.md records only what the user confirms they have set up

Tone of Voice

  • Address the user as "you": "Your weekly update runs on..." not "The user's weekly update runs on..."
  • Avoid hyperbole and cinema poster phrasing (not "set it and forget it" or "your job search on autopilot")
  • Use the Oxford comma (serial comma: "tracker, follow-ups, and market map")
  • Never use em dashes. Use commas, semicolons, colons, or full stops instead

Template Usage

When a capability names a template, load it with the @ reference before writing, and follow its structure and footer. Users and other skills rely on the same headings appearing in the same places.


Related Skills

  • /career-navigator: The application tracker the standup reads
  • /market-mapper: The market map the weekly update refreshes
  • /linkedin-coach: The content calendar the posting reminder reads
  • /getting-started: Orientation; routes here for anything about scheduling

Career Routines | Career Helper Plugin | Prosper AI Consulting, UK

Bundled files

The model reads these on demand while the skill is loaded. They are exposed as readable files and are never executed.

Frequently asked questions

What does the Career Routines AI skill do?

Use when the user wants parts of their job search to run on a schedule, or asks about routines they already have, such as putting the weekly update on a Cowork schedule, setting up narrower recurring tasks, or working out why a scheduled task did not run. Explains when a task runs in the cloud and when it needs their computer.

Why use Career Routines on TypingMind?

Because you install it once and use it with any model. Career Routines is plain Markdown rather than provider-specific code, so the same skill runs on GPT-5, Claude, Gemini, Grok, or a local model — and you can switch model mid-chat without it breaking. TypingMind runs on your own API keys, so you pay providers directly instead of a per-seat subscription, and your skills and chats stay in your own storage.

How do I install Career Routines in TypingMind?

Open Plugins → Skills → Install from GitHub in TypingMind and paste https://github.com/Zal4DW/career-helper/tree/main/career-helper/skills/career-routines. TypingMind reads its SKILL.md and bundles its files and installs it as a skill you can enable per chat.

Which AI models can use Career Routines?

Any model you connect in TypingMind. AI skills are plain Markdown instructions rather than provider-specific code, so GPT, Claude, Gemini, Grok, and local models can all load this skill when a request matches it.

How many AI models can I use with Career Routines?

As many as you like. As long as a model supports skills, you can use Career Routines with it — GPT, Claude, Gemini, Grok, DeepSeek, Mistral, Llama and more — all on TypingMind with your own API keys.

Is the Career Routines AI skill free?

It is published on GitHub by Zal4DW. Check the repository for licensing terms. You only pay your own AI provider for the tokens you use.

What are AI skills?

An AI skill is a reusable instruction bundle that teaches an AI model how to do one specific task. It follows the open Agent Skills format: a SKILL.md file with a name and description, plus any scripts, templates or reference files the model may need. The model reads the instructions only when your request matches the skill, so an installed skill costs nothing until it is used.

How are AI skills different from plugins or MCP servers?

A plugin or MCP server gives a model new tools to call — code that runs somewhere and returns a result. An AI skill gives the model knowledge and process instead: how to approach a task, which steps to follow, what good output looks like. Skills are plain Markdown, so they need no server, no API key and no runtime, and they work with any model.

View all

Set up your own AI workspace now

Get notified about new features and future giveaways by subscribing to our newsletter 👇