redhat-developer / redhat-developer/abbenay
feat(grpc): add tool approval flow for Chat/SessionChat RPCs
Nobody has claimed this yet.
- Dominant language
- TypeScript
- Stars
- 12
- Forks
- 7
- Avg merge
- 7h 7m
- Merged PRs (30d)
- 3
Description
Summary
The REST API supports an interactive tool approval flow for chat:
POST /api/chatstreams SSE events includingapproval_requestwhen a tool invocation needs human approvalPOST /api/chat/:chatId/approveresolves the pending approval (approve or deny)
The gRPC Chat and SessionChat RPCs support streaming responses but have no equivalent approval mechanism. This means gRPC clients cannot replicate the dashboard's interactive chat-with-approval UX.
Proposal
Design a gRPC-native approval flow. Options include:
- Bidirectional streaming — extend
Chat/SessionChatto accept client-to-server approval messages mid-stream - Separate approval RPC — add an
ApproveToolCall(chatId, toolCallId, approved)unary RPC that resolves pending approvals, mirroring the REST endpoint - Callback pattern — use the existing
VSCodeStreambidirectional channel to relay approval requests/responses
Option 2 is simplest and most consistent with the REST pattern.
Context
Identified during gRPC config parity audit in #32.
Contributor guide
No contributing guide indexed for this repository
First steps
- Read the whole issue, then the project's contributing guide.
- Comment on the issue to say you are picking it up — it saves two people doing the same work.
- Fork the repository and make your change on a branch.
- Open a pull request that references the issue number.
Research direction
Start by comparing the gRPC Chat and SessionChat RPC definitions and streaming implementations with the REST approval flow at POST /api/chat and POST /api/chat/:chatId/approve. Review the existing VSCodeStream bidirectional channel and the context from #32 before choosing an approach. Done means a documented gRPC approval design that lets clients receive approval requests and approve or deny pending tool calls.
Written by the indexing model from the issue text.
Assessment
- Tech stack
- grpc, typescript
- Domain
- api, backend
- Issue type
- Feature
- Difficulty
- 5/5
- Estimated time
- Over a week
- Activity status
- Quiet
- Clarity
- Needs clarification
- Newbie friendliness
- 30/100