StackSherpa
@astein91
Intelligent API recommendation engine that silently picks the best APIs for your stack. Analyzes your project profile, compliance requirements, past decisions, and known issues to recommend the right provider for auth, payments, email, storage, and 25+ other categories.
Overview
What is StackSherpa?
StackSherpa is an intelligent API recommendation engine providing detailed profiles for over 450 providers across 28 categories, including pricing, known issues, compliance data, and live GitHub issue tracking. It is designed for developers and teams who need to quickly evaluate and select API providers. The catalog is a shared read-only database accessible via CLI or web, with no API keys required for users.
How to use StackSherpa?
Install the CLI globally with npm install -g stacksherpa, optionally install the Claude Code skill to have Claude automatically consult StackSherpa. Run commands like stacksherpa providers email --pretty to list providers, stacksherpa profile --set to configure project constraints, and stacksherpa decide to log API decisions. All commands output JSON; add --pretty for human-readable formatting.
Key features of StackSherpa
- CLI with commands for categories, providers, profiles, and decisions
- Claude Code skill for automatic API recommendations inside Claude
- Project profiles for personalized provider suggestions
- Live pricing and known issues scraped from provider websites and GitHub
- Local data storage; user data never leaves the machine
Use cases of StackSherpa
- A startup evaluating email APIs can compare providers by pricing, compliance, and known issues.
- A developer integrating a database can quickly list options and see self-hostable flags.
- A team using Claude Code can automatically receive provider recommendations before writing integration code.
- A maintainer can audit and approve newly discovered providers via cron jobs.
FAQ from StackSherpa
Do I need an API key to use StackSherpa?
No. The catalog is a shared read-only database; no API keys are required to browse or query providers.
How does the Claude Code skill work?
Once installed, Claude silently runs stacksherpa providers <category> before suggesting any API integration. No slash command is needed.
Where is my project data stored?
Your project profile and decision history are stored locally in .stacksherpa/ within your project. Global defaults live in ~/.stacksherpa/. No data leaves your machine.
Can I use StackSherpa without the Claude skill?
Yes. You can run the CLI directly or add instructions to your CLAUDE.md to use stacksherpa providers <category> before choosing an API.
How is the catalog maintained?
An automated pipeline scrapes GitHub issues, pricing, and provider details on a regular schedule. Maintainers can approve or reject discovered providers via CLI commands.
