agents
Use when designing, deploying, or debugging a Butterbase Agent (declarative LLM/tool graph), registering an MCP server for tool use, or wiring access controls…
Use as stage 1 of the Butterbase journey, when the user has only a rough idea ("I want to build something that..."). Conducts a concrete, one-question-at-a-time brainstorm that surfaces who the user is, what they do first, what the must-haves are, and inline-tags Butterbase
$ npx -y skills add butterbase-ai/butterbase-skills --skill journey-idea --agent claude-codeHow it fires
How this skill gets triggered: by you, by Claude, or both.
/journey-ideaContext preview
The summary Claude sees to decide when to auto-load this skill.
Use as stage 1 of the Butterbase journey, when the user has only a rough idea ("I want to build something that..."). Conducts a concrete, one-question-at-a-time brainstorm that surfaces who the user is, what they do first, what the must-haves are, and inline-tags Butterbase
name: journey-idea
description: Use as stage 1 of the Butterbase journey, when the user has only a rough idea ("I want to build something that..."). Conducts a concrete, one-question-at-a-time brainstorm that surfaces who the user is, what they do first, what the must-haves are, and inline-tags Butterbase capabilities (→ manage_schema, → deploy_function, etc.). Produces docs/butterbase/01-idea.md.Stage 1 of the guided journey. Turn a vague idea into a written brief with capability tags.
None required. Optionally reads any existing `01-idea.md` to resume.
Follow the cross-cutting rules from the spec:
Walk through these questions in order. Do not skip — but adapt the wording.
1. `"In one sentence, what is the painful thing your user does today that this replaces?"` 2. `"Who is the first user? ① solo creator ② small team ③ end customers of a business ④ Other"` 3. `"What is the first action they take in the app?"` 4. `"If we could ship only ONE screen, which one?"` 5. `"Do two different users see different data when they log in? (yes → user-scoped data → manage_rls; no → shared → simpler)"` 6. `"Are there files/images involved? (yes → manage_storage; no → skip)"` 7. `"Does anything happen on a schedule or in response to an external event? (yes → deploy_function cron/webhook; no → skip)"` 8. `"Is there a knowledge base / search-over-documents feature? (yes → manage_rag_content; no → skip)"` 9. `"Does the UI need live updates or presence? (yes → manage_realtime; no → skip)"` 10. `"Any chat rooms / multiplayer / per-user-actor state? (yes → manage_durable_objects; no → skip)"` 11. `"Any LLM features (chat, embeddings, summarisation)? (yes → manage_ai; no → skip)"` 12. `"Any multi-step LLM+tool workflows the user triggers (e.g. 'summarise my orders', 'auto-tag a doc', 'research and reply')? (yes → create_agent; no → skip)"`. If yes, also ask: `"Will the agent need external MCP servers (e.g. GitHub, Slack, custom MCP)?"` and `"Visibility: ① only the developer ② any logged-in user ③ unauthenticated public — note: public + write tools requires explicit safety acknowledgement."` 13. `"Are you planning to publish this app as a public template others can clone? (yes / unlisted-public / no)"`. Set `publish_as_template: <yes|unlisted|no>` in `00-state.md` front-matter. Public templates require a README with clone instructions and `butterbase repo push` to sync the source tree — flag this so `plan` and the publish stage know to plan for it.
If `hackathon_mode: true` in `00-state.md`: after question 4, also ask:
# Idea **One-liner:** <user's answer to Q1> **First user:** <Q2> **First action:** <Q3> **One screen:** <Q4> ## Must-haves - <feature> → <capability tag> - ... ## Post-hackathon (nice-to-have) - ... ## Capability map | Capability | Used? | Why | |---|---|---| | manage_schema | yes | core data model | | manage_rls | yes | user-scoped data | | manage_oauth | yes | Google sign-in | | manage_storage | no | no files | | deploy_function | yes | daily-digest cron | | manage_ai | no | | | manage_rag_content | no | | | manage_realtime | no | | | manage_durable_objects | no | | | create_agent | no | | | publish_as_template | no | set true if user plans to share this app publicly |
At the end of the idea write-up, also note in one line that the build will use:
This sets expectations before `plan` makes specific package choices.
Claude Code plugin for Butterbase — the AI-Native Backend-as-a-Service. This plugin gives Claude deep knowledge of Butterbase's 42+ MCP tools, guides you through common workflows, and auto-configures the MCP server connection.
Repo: butterbase-ai/butterbase-skills
Use when designing, deploying, or debugging a Butterbase Agent (declarative LLM/tool graph), registering an MCP server for tool use, or wiring access controls…
Use when calling the app's AI gateway from agent tools — chat completions, embeddings, listing models, configuring defaults or BYOK, reading token/cost usage
Use when configuring OAuth providers (Google/GitHub/Apple/X/etc.), setting up post-login auth hooks, tuning JWT lifetimes, or generating service API keys
Use when building a new Butterbase app from scratch, creating a full-stack application, or when the user asks to set up a complete backend with database, auth,…
Use when contributing to the Butterbase codebase, adding new MCP tools, creating API routes, writing migrations, or understanding the monorepo architecture
Use when users report access denied errors, see wrong data, RLS policies are not working, or when troubleshooting Row-Level Security issues in Butterbase