#17845: feat(onboard): add built-in Fireworks provider
docs
cli
commands
agents
stale
size: L
Cluster:
Provider Support Enhancements
## Summary
- add Fireworks as a built-in provider in onboarding/auth flows (interactive + non-interactive)
- wire Fireworks auth/env resolution and implicit provider injection (`FIREWORKS_API_KEY`)
- add docs for provider catalog and wizard/onboard examples, including a new `/providers/fireworks` page
- add e2e coverage for auth-choice mapping, onboarding, CLI flag wiring, and provider resolution
## Testing
- pnpm check
- pnpm exec vitest run -c vitest.e2e.config.ts src/agents/models-config.providers.fireworks.e2e.test.ts src/commands/auth-choice.e2e.test.ts src/commands/onboard-non-interactive.provider-auth.e2e.test.ts src/cli/program.smoke.e2e.test.ts src/commands/onboard-auth.e2e.test.ts src/commands/auth-choice-options.e2e.test.ts src/agents/model-auth.e2e.test.ts
<!-- greptile_comment -->
<h3>Greptile Summary</h3>
Adds Fireworks AI as a built-in model provider with full onboarding support, following the established pattern for OpenClaw provider integrations. The implementation includes interactive and non-interactive CLI flows, environment variable resolution (`FIREWORKS_API_KEY`), implicit provider injection, comprehensive e2e test coverage, and documentation across the provider catalog and wizard examples.
Key changes:
- Added 17 Fireworks models to the catalog (Kimi K2.5 as default, plus GLM-5, DeepSeek V3.2, MiniMax M2.5, and others)
- Wired `fireworks-api-key` auth choice throughout onboarding, auth-choice mapping, and CLI flag system
- Implemented implicit provider injection when `FIREWORKS_API_KEY` env var is set
- Added e2e tests for auth-choice, onboarding (interactive + non-interactive), provider resolution, and CLI flag wiring
- Created `/providers/fireworks` documentation page with config examples
<h3>Confidence Score: 5/5</h3>
- This PR is safe to merge with no blocking issues
- The implementation follows the established provider integration pattern precisely, matches existing providers (Together, Venice, Xiaomi) in structure and scope, includes comprehensive test coverage across all integration points, and maintains consistency with the codebase's conventions
- No files require special attention
<sub>Last reviewed commit: 32cd89e</sub>
<!-- greptile_other_comments_section -->
<sub>(2/5) Greptile learns from your feedback when you react with thumbs up/down!</sub>
<!-- /greptile_comment -->
Most Similar PRs
#15991: feat: add Novita AI provider support with dynamic model discovery
by Alex-wuhu · 2026-02-14
74.9%
#7272: Models: add SiliconFlow provider
by qychen2001 · 2026-02-02
74.7%
#21520: feat(onboard): add Cencori provider.
by bolaabanjo · 2026-02-20
73.8%
#14393: feat: add standalone DeepSeek provider support
by osoulmate · 2026-02-12
73.8%
#8783: feat(provider/volcengine): support volcengine provider
by ShanyouYu-Sean · 2026-02-04
73.7%
#13295: feat: add Eternal AI provider integration
by peterparkernho · 2026-02-10
72.8%
#7418: feat (amazon-nova): add Amazon Nova 1P provider
by 5herlocked · 2026-02-02
72.7%
#19841: feat: add Amazon Bedrock as first-class onboarding provider
by elbeyf · 2026-02-18
72.0%
#15742: feat: add Edgee AI Gateway as provider
by manthis · 2026-02-13
71.9%
#11151: feat: add DeepSeek provider support
by MackDing · 2026-02-07
71.9%