AgentStack
Browse Sign in
Browse Why AgentStack Sell Docs
Sign in
SKILL verified MIT Self-run

Journey Preflight

skill-butterbase-ai-butterbase-skills-journey-preflight · by butterbase-ai

Use before any Butterbase platform-touching action (or when journey-preflight is invoked directly) to verify the user has a Butterbase account, the MCP server is connected, BUTTERBASE_API_KEY is set, and an app_id exists for this project. Re-run automatically if docs/butterbase/03-preflight.md is older than 24h or any required value is null.

No reviews yet
0 installs
13 views
0.0% view→install

Install

$ agentstack add skill-butterbase-ai-butterbase-skills-journey-preflight

✓ scanned · ✓ verified, works with Claude Code, Cursor, and more.

Security review

✓ Passed

No issues found. Passed automated security review. · v0.1.0 How review works →

  • Prompt-injection patterns
  • Secret / credential exfiltration
  • Dangerous shell & filesystem operations
  • Untrusted network calls
  • Known-malicious package signatures

What it can access

  • Network access No
  • Filesystem access No
  • Shell / process execution No
  • Environment & secrets No
  • Dynamic code execution No

From automated source analysis of v0.1.0. “Used” means the capability is present in the source — more access means more to trust, not that it’s unsafe.

View the full security report →

Verified badge

Passed review? Show it. Paste this badge into your README, it links to the public security report.

AgentStack Verified badge Links to your public security report.
[![AgentStack Verified](https://agentstack.voostack.com/badges/verified.svg)](https://agentstack.voostack.com/security/report/skill-butterbase-ai-butterbase-skills-journey-preflight)

Reliability & compatibility

Security review passed
0 installs to date
no reviews yet
1mo ago

Declared compatibility

Claude CodeClaude Desktop

Compatibility is declared by the source manifest. End-to-end runtime verification is coming, see below.

Preview Execution monitoring

We're building live execution health for every listing: tool-call success rate, median latency, uptime, and last-checked timestamps, measured, not self-reported. It isn't live yet, so we don't show numbers we can't stand behind.

How agent discovery & health will work →
Are you the author of Journey Preflight? Claim this listing to set pricing, connect Stripe payouts, and keep 70% of every sale.
Sign up to claim

About

Butterbase Preflight

Gate that ensures the user can actually talk to the Butterbase platform before any stage tries to. Writes results to docs/butterbase/03-preflight.md and stamps app_id + api_base into 00-state.md.

When to use

  • Automatically as the first step of any journey-* build stage when 03-preflight.md is missing, older than 24 hours, or 00-state.md's app_id is null.
  • Directly via /butterbase-skills:journey-preflight when the user wants to re-verify.

Procedure

For each check, write a one-line result to 03-preflight.md ( ✓/✗ ). Ask the user to fix any ✗ before continuing — do not auto-skip failures.

  1. Butterbase account. Call mcp__butterbase__list_regions (cheapest call). Success → ✓. Auth error → ✗ with "Sign up at https://butterbase.ai, then run npx @butterbase/cli mcp install and complete the OAuth flow in your client (e.g. /mcp in Claude Code).". If MCP tool is not visible at all, jump to step 2.
  1. MCP server connected. Check whether mcp__butterbase__* tools are listed in the agent's available tools. Absent → ✗ with "Install the Butterbase MCP server: npx @butterbase/cli mcp install. See https://butterbase.ai/docs/mcp.". Present → ✓.
  1. MCP OAuth complete. Auth for mcp__butterbase__* calls is OAuth, not an env var — step 1 already proved it works if list_regions succeeded. If step 1 failed with auth error, retry after the user runs /mcp (or claude mcp login butterbase) and completes browser consent. BUTTERBASE_API_KEY (bb_sk_…) is still relevant for runtime app code (functions, server-side @butterbase/sdk) — check separately at deploy time, not here.
  1. App provisioned. Read 00-state.md front-matter for app_id.
  • Non-null and mcp__butterbase__manage_app (action get_config) succeeds → ✓.
  • Null → ask the user: "No app yet. Create one now? (yes/no — if no, you can paste an existing app_id)". On yes, call mcp__butterbase__init_app (ask for app name and region — recommend us-east-1 as default). Capture app_id and api_base from the response. Update 00-state.md front-matter.
  • User-provided app_id → verify with manage_app get_config; on success update 00-state.md.
  1. CLI installed. Run butterbase --version. If "command not found":
  • Read the Toolchain section of 02-plan.md. If CLI usage: yes (default), prompt: "Install @butterbase/cli globally now? (yes / skip)". On yes, run npm install -g @butterbase/cli. On skip, note in 03-preflight.md that the CLI is not installed and which workflows will require dashboard fallback.
  • If CLI usage: no, skip this check.

03-preflight.md format

# Preflight — 

- account        ✓
- mcp_connected  ✓
- api_key        ✓
- app_provisioned ✓  app_id=app_abc123 region=us-east-1

Re-run with /butterbase-skills:journey-preflight.

Outputs

  • Writes/overwrites docs/butterbase/03-preflight.md.
  • Updates 00-state.md front-matter app_id, api_base, last_updated, and ticks - [x] preflight.

Anti-patterns

  • ❌ Echoing the API key value back to the user.
  • ❌ Skipping a failed check ("I'll just try anyway"). Stop and ask the user to fix it.
  • ❌ Calling init_app without confirming with the user — apps cost money on paid plans.

Source & license

This open-source skill is cataloged on AgentStack and links to its original source — we do not rehost the code.

Install and usage instructions live in the source repository linked above.

Reviews

No reviews yet, be the first.

Versions

  • v0.1.0 Imported from the upstream source.