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

Accept a JSON Schema document as `output_type`

已关闭
  • 哪里不对？
#5,344 1 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
4/5
预计耗时
3-5 天
新手友好度
25/100
Issue 类型
功能
描述清晰度
基本清楚
活跃度
活跃
技术栈
python
领域
ai

调研方向

Start from AgentOutputSchema and the AgentOutputSchemaBase interface it implements, specifically json_schema(), is_strict_json_schema() and validate_json(), plus how ModelBehaviorError is raised during output parsing. Decide the validation approach first, since the issue leaves open whether to add jsonschema as an optional extra or skip local validation when strict=True. Done means a JsonSchemaOutput class that passes its schema through unchanged when strict=False, returns a parsed dict, and raises ModelBehaviorError on invalid JSON or schema violations, with tests for both modes.

由索引模型根据 Issue 内容生成。

描述

Please read this first
  • Have you read the docs? Yes
  • Have you searched for related issues? Yes
Describe the feature

AgentOutputSchema takes a Python type, so an agent whose response schema arrives at runtime as JSON Schema (user-defined agents in agent-builder products, schemas stored in config) needs a hand-written AgentOutputSchemaBase subclass that re-implements json_schema(), is_strict_json_schema() and validate_json(), with its own JSON parsing, validation and ModelBehaviorError mapping.

Proposal: a built-in output schema that takes the document directly, e.g.

Agent(..., output_type=JsonSchemaOutput(schema, name="ticket", strict=False))
  • strict=True runs ensure_strict_json_schema as AgentOutputSchema does; strict=False sends the schema as given, since user-authored schemas often fall outside the strict subset.
  • validate_json parses and validates against the schema and raises ModelBehaviorError on invalid JSON or a schema violation, so retries and error handling match typed outputs.
  • It returns the parsed dict.

Validation could use jsonschema as an optional extra, or skip local validation when strict=True, where the provider already enforces the schema.

主要语言
Python
星标
30k
派生
4.9k
平均合并
16 小时 25 分钟
30 天内合并 PR
220

环境准备

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。

openai/openai-agents-python 的其他 Issue

查看 openai/openai-agents-python 的全部 Issue

相似的 Issue

更多 Python Issue

把新 issue 发到你的邮箱

精选适合新手参与的 GitHub issue 摘要。