Hacktoberfest 2026 : les issues que les mainteneurs ont marquées pour octobre, ouvertes et accessibles aux débutants. Parcourir les issues Hacktoberfest

[bot] Instrument Azure AI Agents SDK (204,195 weekly downloads)

Fermée
  • Quelque chose ne va pas ?
#2,605 0 commentaires 0 réactions 0 personnes assignées Voir sur GitHub

Les mainteneurs répondent en général sous 1 jour

Personne n'a encore pris cette issue.

Évaluation

Difficulté
4/5
Temps estimé
3-5 jours
Accessibilité débutants
48/100
Type d'issue
Fonctionnalité
Clarté
Plutôt claire
Activité
Active
Stack technique
azure, javascript, typescript
Domaine
devtools, observability

Piste de recherche

Start by reading js/src/auto-instrumentations/configs/all.ts and js/src/span-origin.ts, then compare existing SDK instrumentation with the Azure AI Agents create, thread, run, and streaming APIs described in the linked upstream docs. The issue identifies e2e/scenarios/* as the place to check coverage. Done means the agent execution surface is instrumented and represented in the registry and end-to-end scenarios.

Rédigé par le modèle d'indexation à partir du texte de l'issue.

Description

typescript

Summary

@azure/ai-agents is Microsoft's official TypeScript/JavaScript SDK for Azure AI Foundry's Agent Service — a stateful, server-run agent surface (analogous to OpenAI's Assistants/Responses agent model) where you create an agent, create a thread, and execute runs that the service drives end-to-end (including tool calls such as file search and code interpreter), optionally streamed. This execution surface is not instrumented anywhere in this repository. This is distinct from Azure OpenAI/Azure AI Foundry model-routing support, which Braintrust already documents as an AI provider (gateway/billing config), not a client-side agent-run tracing integration.

What's missing

  • AgentsClient.createAgent(...) — registers an agent (model + instructions + tools).
  • Thread + run execution (client.threads.create(), client.runs.create()/createAndRun, streaming runs) — the actual agent-execution surface, where the service plans, calls the model, and executes tool calls (file search, code interpreter, function tools) before returning.
  • The higher-level @azure/ai-projects client, which wraps @azure/ai-agents behind a .agents property for the same create/run/stream surface.

None of @azure/ai-agents, @azure/ai-projects, AgentsClient, or azure-ai-agents appear anywhere in js/src/, js/src/auto-instrumentations/configs/all.ts (the instrumentation registry), js/src/span-origin.ts (INSTRUMENTATION_NAMES), or any e2e/scenarios/* directory.

Weekly downloads

Weekly downloads: 219,062 (as of 2026-10-12; window 2026-10-04–2026-10-10; source https://api.npmjs.org/downloads/point/last-week/@azure/ai-agents)

Braintrust docs status

not_found — Checked https://www.braintrust.dev/docs/integrations/ai-providers/azure, which covers Azure OpenAI/Azure AI Foundry as a model provider (chat completions via AzureOpenAIClient, Entra Workload Identity auth for the Braintrust gateway) — no mention of the Agents/Agent Service SDK, threads, or runs. Also checked https://www.braintrust.dev/docs/guides/traces/integrations (full agent-framework integration list: Agno, OpenAI Agents SDK, Claude Agent SDK, Pi Coding Agent SDK, Deep Agents, LangGraph, CrewAI, AutoGen, AgentScope, Google ADK, LiveKit Agents, Mastra, Pipecat, Pydantic AI, Strands Agents SDK, Cloudflare Agents/AI Chat) — Azure AI Agents is absent.

Upstream sources

Local repo files inspected

  • js/src/auto-instrumentations/configs/all.ts (full instrumentation registry — no Azure Agents entry)
  • js/src/span-origin.ts (INSTRUMENTATION_NAMES — no Azure Agents entry)
  • Repo-wide case-insensitive search for @azure/ai-agents, @azure/ai-projects, AgentsClient, azure-ai-agents across the working tree (zero matches)
  • e2e/scenarios/ directory listing (no Azure-Agents-named scenario)
Langage dominant
TypeScript
Étoiles
29
Forks
16
Merge moyen
1 j 22 h
PR mergées (30 j)
72

Préparer son environnement

  • Fournit un Dockerfile ou un fichier Docker Compose
  • Aucun modèle de pull request
  • Aucun guide de contribution

Par où commencer

  1. Lisez l'issue en entier, puis le guide de contribution du projet.
  2. Signalez en commentaire que vous la prenez — cela évite que deux personnes fassent le même travail.
  3. Forkez le dépôt et travaillez sur une branche.
  4. Ouvrez une pull request qui référence le numéro de l'issue.

Autres issues de braintrustdata/braintrust-sdk-javascript

Toutes les issues de braintrustdata/braintrust-sdk-javascript

Issues similaires

Plus d'issues TypeScript

Recevez les nouvelles issues par e-mail

Un résumé court des issues GitHub adaptées aux débutants.