/install fulcra-onboarding
Fulcra Onboarding
Primary onboarding flow for new users connecting to the Fulcra environment. Fulcra is a universal data and memory backend for agents that securely stores personal context and custom data streams.
General Guidelines
- Tone & Vibe (The "Wow" Factor): Fulcra unlocks massive potential because it acts as a universal memory and data backend for agents. Your tone should reflect this excitement and vast possibility. Be engaging, conversational, and fun.
- Use Emojis: Liberally (but tastefully) use emojis throughout the onboarding process, especially when presenting lists, options, or celebrating a successful step (e.g., "Would you like to track ☕ Coffee, 📚 Books, or 🏃 Fitness?").
- Maintain Momentum: Keep the user moving forward. If they get stuck, make a choice for them or offer a heavily opinionated default to get them to the "Time-to-Wow" demonstration as quickly as possible.
Workflow
-
Pre-flight Check:
- Read and execute the instructions in
references/fulcra-onboarding-prerequisites.mdto silently verify and installuvif necessary. This must pass before proceeding.
- Read and execute the instructions in
-
Discovery:
- Read and execute the instructions in
references/fulcra-onboarding-discovery.mdto handle user authentication, uncover their core intent, and suggest concrete use cases. Wait for this phase to hand back 2-3 specific custom data types/streams before proceeding.
- Read and execute the instructions in
-
Data Modeling:
- Translate the user's intent into 2-3 specific custom data types/streams (Annotations).
- Read and follow the instructions in
references/fulcra-onboarding-create-annotations.mdto define and create these schemas. - Crucial Memory Step: When you create these annotations, explicitly remember the returned
ANNOTATION_IDand the exactdata_typefor each one. You must use these IDs directly in the next step—do not make unnecessary API calls to look them up again.
-
Record First Data:
- Pick one of the newly created annotations and ask the user a direct question to get their first piece of data (e.g., if you created a "coffee consumed" annotation, ask "How many coffees have you had today?").
- Once they answer, record their response into Fulcra by reading and following the instructions in
references/fulcra-onboarding-record-annotations.md.
-
Time-to-Wow (The Demonstration):
- Read and execute the instructions in
references/fulcra-onboarding-demonstration.mdto retrieve the recorded data, ask the user for a preferred aesthetic, and generate a custom themed HTML dashboard to display directly in the chat.
- Read and execute the instructions in
-
Handoff & Next Steps:
- App Download (Required): Always conclude by directing the user to the Fulcra Context iOS app.
- Value Proposition: Briefly explain that the app unlocks automatic background sync (Apple Health, location, calendar) and gives them a quick, native interface to log the custom Annotations they just created.
- Optional Momentum: If the user is highly engaged, suggest running their new dashboard locally or exploring further bulk data integrations.
- Make sure OpenClaw is installed (local or Docker)
- Run the install command in chat:
/install fulcra-onboarding - After installation, invoke the skill by name or use
/fulcra-onboarding - Provide required inputs per the skill's parameter spec and get structured output
What is Fulcra Onboarding?
Guides a new user or agent through the initial setup, configuration, and capabilities of the Fulcra environment. It is an AI Agent Skill for Claude Code / OpenClaw, with 39 downloads so far.
How do I install Fulcra Onboarding?
Run "/install fulcra-onboarding" in the OpenClaw or Claude Code chat to install it in one step — no extra setup required.
Is Fulcra Onboarding free?
Yes, Fulcra Onboarding is completely free, licensed under MIT-0. You can download, install and use it at no cost.
Which platforms does Fulcra Onboarding support?
Fulcra Onboarding is cross-platform and runs anywhere OpenClaw / Claude Code is available (cross-platform).
Who created Fulcra Onboarding?
It is built and maintained by Fulcra.ai (@fulcra); the current version is v0.0.1.