run-llama/LlamaIndexTS

Usage Visibility for Anthropic and OpenAI when using agent-workflow

Open

#2,207 opened on Sep 18, 2025

View on GitHubContributor guide
 (2 comments) (0 reactions) (0 assignees)TypeScript (524 forks)auto 404
buggood first issuehelp wanted

Repository metrics

Stars
 (3,078 stars)
PR merge metrics
 (PR metrics pending)

Description

1. Anthropic (Streaming)

When using agent-workflow with Anthropic and stream = true, the usage info is filtered out from the chunk.
The usage info is of type "message-delta", and the chunk is not yielded in the code below:

🔗 LlamaIndexTS Anthropic LLM Code

Example payload:

{
  "type": "message_delta",
  "delta": {
    "stop_reason": "end_turn",
    "stop_sequence": null
  },
  "usage": {
    "input_tokens": 8,
    "cache_creation_input_tokens": 0,
    "cache_read_input_tokens": 0,
    "output_tokens": 12
  }
}

2. Anthropic and OpenAI (Non-Streaming)

When running with agent.run, it is returning the finalEvent. However, the finalEvent does not have usage info applicable. https://github.com/run-llama/LlamaIndexTS/blob/main/packages/workflow/src/agent/agent-workflow.ts#L470-L476

Contributor guide

Research direction
Inspect the Anthropic streaming code in `packages/providers/anthropic/src/llm.ts` around line 584-618 to understand why `message delta` chunks are filtered out, and modify the chunk processing to include usage info. For non streaming, examine `packages/workflow/src/agent/agent workflow.ts` around line 470-476 to see how `finalEvent` is constructed and ensure usage data is attached.
Tech stack
typescript
Domain
backend
Issue type
Bug
DifficultyEstimated implementation difficulty for a new contributor, from 1 for very small changes to 5 for expert-level work.
2
Estimated timeA rough time range for an experienced contributor to investigate, implement, test, and prepare a pull request.
1-3 hours
Activity statusHow available the issue appears right now: fresh, active, stale, blocked, or waiting on maintainer input.
Active
ClarityHow clearly the issue explains the expected change, acceptance criteria, and next step.
Clear
Prerequisites
TypeScriptGit
Newbie friendlinessA 1-100 score estimating how approachable this issue is for first-time contributors.
75