Skip to content

PLAN.md

Latest commit

 

History

History
151 lines (111 loc) · 3.82 KB

PLAN.md

File metadata and controls

151 lines (111 loc) · 3.82 KB
Raw

csfloat-node-sdk Plan

Date: 2026-03-07 Scope: create a strong public OSS starting point for a Node.js / TypeScript SDK around the CSFloat API, with honest room for future CS2 market adapters.

1. Core Positioning

This repository should start as:

  1. an unofficial Node.js / TypeScript SDK for the CSFloat API
  2. a public OSS project that is actually useful on its own
  3. a clean foundation for future automation tools and adapters

This repository should not start as:

  1. a vague “all CS2 markets” wrapper without real implementation
  2. a thin copy of another unofficial wrapper
  3. a giant product surface with unclear ownership

2. Repository Goal

Deliver a public repository that can credibly say:

  1. here is a well-designed CSFloat Node.js SDK
  2. here are the tested API workflows it supports
  3. here is the roadmap toward a broader market-adapter architecture

3. Current Scope Freeze

Current implementation scope:

  1. CSFloat API only
  2. Node.js / TypeScript only
  3. explicit support for authenticated account flows
  4. focus on listings, stall, inventory, account profile, and listing mutation

Current non-goals:

  1. browser automation
  2. pricing engine logic
  3. Telegram bot workflows
  4. multi-market support in v0

4. Roadmap Direction

Future direction can expand toward a broader CS2 market tooling ecosystem, but only in stages.

Recommended wording:

  1. current: CSFloat-first SDK
  2. next: market-adapter-ready architecture
  3. future: additional CS2 market adapters

That keeps the repository honest while still showing active growth potential.

5. v0.1.0 Deliverables

The first strong public version should include:

  1. typed API client
  2. auth handling via API key
  3. methods for:
    • getMe
    • getInventory
    • getStall
    • getListings
    • getListingById
    • createListing
    • updateListing
    • deleteListing
  4. pagination helpers
  5. normalized error model
  6. usage examples
  7. endpoint support matrix in docs
  8. basic smoke-test guidance

6. Suggested Package Design

Suggested high-level modules:

  1. src/client/
    • low-level HTTP client
    • auth injection
    • retry / timeout handling
  2. src/resources/
    • account
    • inventory
    • stall
    • listings
    • history
  3. src/types/
    • API response types
    • shared entities
  4. src/errors/
    • typed errors
  5. src/utils/
    • query helpers
    • pagination helpers
  6. examples/
    • real usage scripts

7. Documentation Requirements

The public repository should include:

  1. README.md
  2. quick start
  3. install instructions
  4. auth usage
  5. endpoint examples
  6. supported / partially-supported endpoint table
  7. roadmap section
  8. contribution guidelines later

8. Quality Bar For Public Release

Before public release, the repo should have:

  1. clean naming
  2. consistent TypeScript types
  3. no hardcoded secrets
  4. simple examples that actually run
  5. smoke-tested mutation examples clearly labeled
  6. clear disclaimer that this is unofficial

9. Future Multi-Market Expansion

If the repository later grows beyond CSFloat, it should do so by adapters, not by scope drift.

Good future shape:

  1. keep csfloat-node-sdk strong as the first adapter
  2. later introduce adapter interfaces or sibling packages
  3. only claim broad market coverage after real implementations exist

Possible future branches:

  1. market-adapters abstraction layer
  2. normalized listing model across providers
  3. additional CS2 market connectors
  4. comparison / research utilities

10. Best Immediate Next Step

The best next execution step after this planning repo exists:

  1. initialize the actual package skeleton
  2. choose TypeScript + tsup or equivalent minimal build tooling
  3. implement authenticated client bootstrap
  4. implement getMe, getInventory, getStall, and getListings
  5. document real tested behavior as the source of truth