Hacktoberfest 2026:维护者为十月标记出来的 issue,仍然开放、适合新手。 浏览 Hacktoberfest issue

[bot] Instrument aws-sdk-bedrockagentcore (43K weekly downloads)

已关闭
#205 0 条评论 0 个 reaction 已指派 0 人 在 GitHub 查看

维护者通常 1 天内回复

还没有人认领这个 Issue。

评估

难度
5/5
预计耗时
一周以上
新手友好度
35/100
Issue 类型
功能
描述清晰度
基本清楚
活跃度
冷清
技术栈
aws, ruby

调研方向

首先阅读 lib/braintrust/contrib.rb 以及 lib/braintrust/contrib/ 下现有的集成目录,然后查看链接的 API 参考中 Aws::BedrockAgentCore::Client 的入口点。检查 Appraisals 和 braintrust.gemspec 以了解依赖项和测试设置。当 AgentCore 的执行接口(包括 invoke_agent_runtime 和相关的工具调用)已在 SDK 的 tracing 集成中得到表示时,即视为完成。

由索引模型根据 Issue 内容生成。

描述

Summary

The aws-sdk-bedrockagentcore gem is the official, auto-generated AWS SDK client for Amazon Bedrock AgentCore — AWS's newer (2025-era) managed runtime for deploying and operating AI agents (and non-Bedrock, framework-agnostic agents, e.g. built with LangGraph/CrewAI/Strands) at scale. It exposes a distinct agent-run execution API (invoke_agent_runtime) separate from both raw model invocation and the classic Agents-for-Bedrock API, and is not instrumented by this SDK.

This is a separate gem/service from both aws-sdk-bedrockruntime (tracked in #152 — raw converse/invoke_model calls) and aws-sdk-bedrockagentruntime (classic Bedrock Agents invoke_agent, a distinct gem/API from this one). AgentCore Runtime is purpose-built to host and invoke arbitrary agent code (any framework, any model provider) as a managed session-based execution boundary — a clear, high-value agent-run tracing surface for a Ruby SDK whose stated purpose includes agent-execution observability.

What is missing

No instrumentation exists for any aws-sdk-bedrockagentcore execution surface. Key API:

  • client.invoke_agent_runtime(agent_runtime_arn:, runtime_session_id:, payload:, qualifier:, content_type:, accept:, mcp_session_id:, mcp_protocol_version:, ...) — Invokes a deployed AgentCore Runtime (the hosted agent) for one turn within a session, returning the agent's response payload (supports both buffered and streaming/SSE responses). This is the primary execution entry point for any agent hosted on AgentCore, regardless of the underlying agent framework or model provider — the direct agent-run analog to this SDK's existing model-call instrumentation.
  • Related generative/agent surfaces on the same client: invoke_code_interpreter (executes code in an AgentCore-managed sandbox) and invoke_browser (drives an AgentCore-managed browser tool) — both are tool-execution calls typically made during an agent run.
What a span should capture
  • Input: payload (the agent request body), runtime_session_id, agent_runtime_arn, qualifier (endpoint/version)
  • Metadata: provider (aws-bedrock-agentcore), content_type, mcp_session_id/mcp_protocol_version when present (AgentCore Runtime supports exposing agents as MCP servers)
  • Output: response payload, content_type/accept of the response, streaming status
  • Child spans: invoke_code_interpreter/invoke_browser calls made mid-run, mirroring how this SDK already models tool-call child spans for other integrations

Weekly downloads

43K (as of 2026-08-10; https://clickgems.clickhouse.com/dashboard/aws-sdk-bedrockagentcore) — ClickGems reports Last day / Last week / Last month / Total downloads sourced from RubyGems.org's public download dataset; RubyGems.org itself only exposes total and per-version download counts, not a weekly figure, so ClickGems was used as the directly-reported weekly stat.

Additional signal: current version 1.46.0 (per https://rubygems.org/api/v1/gems/aws-sdk-bedrockagentcore.json, checked 2026-08-10), part of the AWS SDK for Ruby's continuous auto-generated release train; total downloads 2,273,730 — notably high relative to the service's 2025 launch, indicating fast-growing adoption.

Braintrust docs status

not_found — Checked https://www.braintrust.dev/docs/integrations/sdk-integrations (Ruby-relevant tracing integrations list only OpenAI, Anthropic, Ruby LLM/ruby-openai; the separate "Agent frameworks" section lists OpenAI Agents SDK, Claude Agent SDK, LangGraph, CrewAI, AutoGen, Google ADK, Strands Agents SDK, Mastra, Pydantic AI, etc. — no Amazon Bedrock AgentCore entry) and https://www.braintrust.dev/docs generally. No mention of Bedrock AgentCore, BedrockAgentCore, or invoke_agent_runtime anywhere in Braintrust's documentation.

Upstream sources

Braintrust docs sources checked

Local repo files inspected

  • lib/braintrust/contrib.rb — registers only 4 integrations: OpenAI, RubyOpenAI, RubyLLM, Anthropic. No AWS/Bedrock integration of any kind.
  • lib/braintrust/contrib/ — contains only openai/, ruby_openai/, ruby_llm/, anthropic/, and rails/ directories. No bedrock or aws directory.
  • Appraisals — no appraisal scenarios for aws-sdk-bedrockagentcore or any AWS gem
  • braintrust.gemspec — no mention of aws or bedrock
  • Grep for bedrock, Bedrock, AgentCore, invoke_agent_runtime across lib/braintrust/ returns zero matches
  • Confirmed issue #152 (aws-sdk-bedrockruntime-not-instrumented) covers only the separate Aws::BedrockRuntime::Client (converse/invoke_model); searched all 46 currently-open issues and found no existing issue mentioning BedrockAgent or AgentCore, so this is a distinct, non-duplicate gap from both #152 and the newly-filed aws-sdk-bedrockagentruntime gap.
主要语言
Ruby
星标
9
派生
11
平均合并
3 天 43 分钟
30 天内合并 PR
9

环境准备

  • 提供 Dockerfile 或 Docker Compose 文件
  • 没有 Pull Request 模板
  • 阅读贡献指南

从这里开始

  1. 先读完整个 Issue,再读项目的贡献指南。
  2. 在 Issue 下留言说明你要接手 —— 这能避免两个人做同样的事。
  3. Fork 仓库,在一个分支上完成修改。
  4. 提交 Pull Request,并在描述里引用这个 Issue 编号。

braintrustdata/braintrust-sdk-ruby 的其他 Issue

查看 braintrustdata/braintrust-sdk-ruby 的全部 Issue

相似的 Issue

更多 Ruby Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。