anomalyco/opencode
1
0
Fork 0
mirror of https://github.com/anomalyco/opencode.git synced 2026-07-23 02:36:57 -04:00
Code Issues 5.9k Packages Projects Releases 19 Wiki Activity

[PR #5022] feat: Add SAP AI Core provider support #11209

New Issue
Closed
opened 2026-02-16 18:16:00 -05:00 by yindo · 0 comments
yindo commented 2026-02-16 18:16:00 -05:00
Owner
Copy Link
Copy Source

Original Pull Request: https://github.com/anomalyco/opencode/pull/5022

State: closed
Merged: No

Summary

This PR adds support for SAP AI Core as a provider in OpenCode, enabling users to access 40+ AI models (GPT-4o, Claude, Gemini, Amazon Nova, etc.) through SAP Business Technology Platform.

Implementation Details

Provider Integration

  • Uses @mymediset/sap-ai-provider (community Vercel AI SDK compatible package)
  • Custom async loader pattern to handle OAuth2 authentication
  • Follows existing OpenCode provider architecture patterns

Configuration

Users can configure SAP AI Core via:

  1. Service Key: SAP_AI_SERVICE_KEY (JSON from SAP BTP Cockpit)
  2. Deployment ID: SAP_AI_DEPLOYMENT_ID (default: d65d81e7c077e583)
  3. Resource Group: SAP_AI_RESOURCE_GROUP (default: default)
  4. Auth Storage: opencode auth add sap-ai-core

Features Supported

  • ✅ OAuth2 authentication with automatic token refresh
  • ✅ 40+ models (OpenAI, Anthropic, Google, Amazon, Meta, Mistral, etc.)
  • ✅ Streaming responses
  • ✅ Tool calling (function calling)
  • ✅ Multi-modal inputs (text + images)
  • ✅ Structured outputs
  • ✅ Model parameters (temperature, topP, maxTokens, etc.)
  • ✅ Data masking via SAP Data Privacy Integration (DPI)

Files Changed

  • packages/opencode/package.json - Added @mymediset/sap-ai-provider@1.3.1
  • packages/opencode/src/provider/provider.ts - Added custom loader for SAP AI Core
  • packages/web/src/content/docs/providers.mdx - Added documentation
  • bun.lock - Updated dependencies

Testing Status

  • ✅ TypeScript compilation passes
  • ✅ Build succeeds (all platforms)
  • ⏳ Manual testing with real SAP AI Core instance (requires maintainer validation)

Documentation

Complete user-facing documentation added to providers.mdx including:

  • Prerequisites (SAP BTP subscription)
  • Service key acquisition steps
  • Configuration examples
  • Supported models list
  • Model parameters guide

Conformity with OpenCode Standards

  • ✅ Follows existing provider architecture (custom loader pattern like github-copilot)
  • ✅ Uses Auth.get() for credential management
  • ✅ Type-safe implementation with proper TypeScript types
  • ✅ Documentation follows exact formatting of existing providers
  • ✅ No style guideline violations

Why This Provider?

SAP AI Core is an enterprise-grade AI platform widely used in SAP ecosystem and beyond. This integration allows:

  • Enterprise users to leverage their existing SAP BTP infrastructure
  • Access to RAG (Retrieval-Augmented Generation) deployments
  • Enterprise compliance (data masking, audit trails)
  • Unified model access across OpenAI, Anthropic, Google, Amazon models

Questions for Maintainers

  1. Should we create a GitHub issue first per CONTRIBUTING.md guidelines?
  2. Any preferred approach for testing without exposing SAP credentials?
  3. Is the default deployment ID acceptable or should we require explicit configuration?

Note: This PR is in draft status to allow for review and validation before final testing.

**Original Pull Request:** https://github.com/anomalyco/opencode/pull/5022 **State:** closed **Merged:** No --- ## Summary This PR adds support for SAP AI Core as a provider in OpenCode, enabling users to access 40+ AI models (GPT-4o, Claude, Gemini, Amazon Nova, etc.) through SAP Business Technology Platform. ## Implementation Details ### Provider Integration - Uses `@mymediset/sap-ai-provider` (community Vercel AI SDK compatible package) - Custom async loader pattern to handle OAuth2 authentication - Follows existing OpenCode provider architecture patterns ### Configuration Users can configure SAP AI Core via: 1. **Service Key**: `SAP_AI_SERVICE_KEY` (JSON from SAP BTP Cockpit) 2. **Deployment ID**: `SAP_AI_DEPLOYMENT_ID` (default: `d65d81e7c077e583`) 3. **Resource Group**: `SAP_AI_RESOURCE_GROUP` (default: `default`) 4. **Auth Storage**: `opencode auth add sap-ai-core` ### Features Supported - ✅ OAuth2 authentication with automatic token refresh - ✅ 40+ models (OpenAI, Anthropic, Google, Amazon, Meta, Mistral, etc.) - ✅ Streaming responses - ✅ Tool calling (function calling) - ✅ Multi-modal inputs (text + images) - ✅ Structured outputs - ✅ Model parameters (temperature, topP, maxTokens, etc.) - ✅ Data masking via SAP Data Privacy Integration (DPI) ### Files Changed - `packages/opencode/package.json` - Added `@mymediset/sap-ai-provider@1.3.1` - `packages/opencode/src/provider/provider.ts` - Added custom loader for SAP AI Core - `packages/web/src/content/docs/providers.mdx` - Added documentation - `bun.lock` - Updated dependencies ## Testing Status - ✅ TypeScript compilation passes - ✅ Build succeeds (all platforms) - ⏳ Manual testing with real SAP AI Core instance (requires maintainer validation) ## Documentation Complete user-facing documentation added to `providers.mdx` including: - Prerequisites (SAP BTP subscription) - Service key acquisition steps - Configuration examples - Supported models list - Model parameters guide ## Conformity with OpenCode Standards - ✅ Follows existing provider architecture (custom loader pattern like `github-copilot`) - ✅ Uses `Auth.get()` for credential management - ✅ Type-safe implementation with proper TypeScript types - ✅ Documentation follows exact formatting of existing providers - ✅ No style guideline violations ## Why This Provider? SAP AI Core is an enterprise-grade AI platform widely used in SAP ecosystem and beyond. This integration allows: - Enterprise users to leverage their existing SAP BTP infrastructure - Access to RAG (Retrieval-Augmented Generation) deployments - Enterprise compliance (data masking, audit trails) - Unified model access across OpenAI, Anthropic, Google, Amazon models ## Questions for Maintainers 1. Should we create a GitHub issue first per CONTRIBUTING.md guidelines? 2. Any preferred approach for testing without exposing SAP credentials? 3. Is the default deployment ID acceptable or should we require explicit configuration? --- **Note**: This PR is in draft status to allow for review and validation before final testing.
yindo added the pull-request label 2026-02-16 18:16:00 -05:00
yindo closed this issue 2026-02-16 18:16:00 -05:00
Sign in to join this conversation.
No Branch/Tag Specified
Branches Tags
Labels
Clear labels
acp
bad
bug
discussion
docs
documentation
enhancement
extension
good first issue
growth
help-wanted
mcp
model-problem
needs:compliance
nix
opentui
perf
pull-request
Mirrored from GitHub Pull Request
 tui
web
windows
zen
No Label pull-request
Milestone
No items
No Milestone
Projects
Clear projects
No project
Assignees
Clear assignees
yindo
No Assignees
1 Participants
Notifications
Due Date
No due date set.
Dependencies

No dependencies set.

Reference: anomalyco/opencode#11209
Delete Branch "%!s()"

Deleting a branch is permanent. Although the deleted branch may continue to exist for a short time before it actually gets removed, it CANNOT be undone in most cases. Continue?