Changelog #
All notable changes to the Nellie API will be documented in this file.
The format is based on Keep a Changelog,
and this project adheres to Semantic Versioning.
[Unreleased] #
Planned #
- Additional output formats
- Custom chapter length controls
- Multi-language content generation
- Batch generation endpoint
- POST
/v1/book— Asynchronous book generation endpoint - Support for 30+ genres/styles
- 9 output types (novel, comic, manga, etc.)
- 6 export formats (txt, pdf, epub, md, html, json)
- AI image generation option
- Custom author attribution
- Custom tone guidance
- Webhook callback support
- GET
/v1/status/{requestId}— Job status polling endpoint - Real-time progress percentage
- Status messages log
- Result URL on completion
- Credit usage tracking
- GET
/v1/configuration— Configuration options endpoint - Available styles list
- Available types list
- Available output formats list
- GET
/v1/models— Available models endpoint - Model descriptions
- Cost per generation
- GET
/v1/usage— Usage statistics endpoint - Total requests count
- Total credits consumed
- Recent request history
- API key authentication via
X-API-Keyheader - Key format:
nel_prefix - Key management in Nellie app dashboard
- Webhook notifications on job completion
- HMAC-SHA256 signature verification
X-Nellie-Signatureheader with timestamp- Replay attack protection (5-minute window)
- Automatic retry on failure (3 attempts)
- Nellie 2.0 (Standard) — 250 credits, balanced performance
- Nellie 3.0 (Premium) — 500 credits, enhanced creativity
- Burst limit: 1 request per 6 seconds
- Daily limit: 15 requests per API key
- Transactional enforcement in Firestore
- Official Python SDK
nellie-api - Type hints and dataclasses
- Automatic retry with exponential backoff
- Webhook signature verification utilities
- Wait-for-completion polling helper
- SSRF protection for webhook URLs
- Webhook signature verification
- API key rotation support
- Environment variable configuration
- API Quick Start Guide
- Full API Documentation
- Python SDK Reference
- Webhook Integration Guide
- Error Reference
- Rate Limits Guide
- Best Practices
[1.0.0] – 2025-12-01 #
Added #
Core API #
Authentication #
Webhooks #
AI Models #
Rate Limiting #
SDK #
Security #
Documentation #
Version History Summary #
| Version | Date | Highlights |
|---|---|---|
| 1.0.0 | Dec 2025 | Initial public release |
Upgrade Notes #
Upgrading to 1.0.0 #
This is the initial release. No upgrade steps required.
For users of the beta/preview API:
- Update base URL to
https://api.nelliewriter.com/v1 - Ensure all endpoints use
/v1/prefix - Update SDK to version 0.1.0+
- Review rate limits (may have changed from beta)
Versioning Policy #
API Versioning #
- Major versions (
/v1/,/v2/) indicate breaking changes - Minor updates are backward compatible within a version
- Deprecated features have minimum 6-month sunset period
- Version in URL path (not headers)
- Follows Semantic Versioning
- Major: Breaking API changes
- Minor: New features, backward compatible
- Patch: Bug fixes, backward compatible
SDK Versioning #
Deprecation Process #
- Announcement — Feature marked deprecated in docs
- Warning — Response headers include deprecation warnings
- Sunset — Feature removed after announced date
Related Documentation #
- Migration Guide — Upgrading between versions
- API Documentation — Current API reference