Week 45, updated 2 Nov 2026

Automation & agent frameworks OpenAI Agents SDK (Python)

Is OpenAI Agents SDK (Python) safe?

Short answer: Caution. StackGrade's weekly scan of openai/openai-agents-python gives it a Caution label for 3 critical-severity known vulnerabilities in runtime dependencies (OSV). It does list 10 high-severity advisories in runtime dependencies, which don't change the label. Critical advisories are counted once per package; a known issue in a locked package version does not prove OpenAI Agents SDK (Python) is exploitable in your setup. Automated scan of commit e355d69 on 2 Nov 2026, not a manual audit.

Updated : the last week its score, rank or safety result changed.

OpenAI's lightweight Python framework for multi-agent workflows.

  • 30,447 stars
  • pushed 1 Nov
  • MIT
  • Python

github.com/openai/openai-agents-python · openai.github.io/openai-agents-python

39score out of 100

#42of 52 overall#12in Automation & agent frameworks▲ 2 vs Week 44

  • Maintenance 30%32
  • Adoption 30%25
  • Docs 20%69
  • Community 20%–
W42W43W44W45

Score over the last 4 weeks: 41, 38, 37, 39.

Caution

3 critical-severity known vulnerabilities in runtime dependencies (OSV).

How this score works

Each part is a percentile among all 52 tracked tools, so 50 is average for this list. Community is not measured yet; the score uses the other 80%.

  • Maintenance: Last push 1 Nov 2026. Last 10 GitHub releases within 65 days (about 4.2 per 30 days), latest release 15 Oct 2026. Median 100.2 h to close an issue (8 closed issues).
  • Adoption: 30,447 stars (70%), 4,970 forks (30%).
  • Docs: README 9,009 characters, 15 headings, 6 code blocks, install section: yes.

Methodology

#12 of 14 in Automation & agent frameworks

Ranked by score. The yellow row is this tool.

Automation & agent frameworks, Week 45score 0–100, safety
  1. 1n8n#3 overall70Caution
  2. 2AutoGPT#14 overall62Caution
  3. 3CrewAI#18 overall57Caution
  4. 4Dify#20 overall56Caution
  5. 5LangChain#23 overall54Caution
  6. 6Pydantic AI#31 overall46Caution
  7. 7Mastra#33 overall44No issues found
  8. 8LangGraph#34 overall44No issues found
  9. 9Agent Development Kit (ADK, Python)#36 overall43Caution
  10. 10OpenBot#39 overall41Caution
  11. 11AutoGen#41 overall40Caution
  12. 12OpenAI Agents SDK (Python)#42 overall39Caution
Show the other 2
  1. 13smolagents#47 overall30No dependencies scanned
  2. 14artemis#50 overall25Caution

Safety scan, 2 Nov 2026

Automated scan of commit e355d69, 1,655 files. Not a manual audit.

Caution

  • 3Critical dependency advisories10 high, 13 medium, 4 low. OSV, runtime dependencies only.
  • 0Install hooksNo npm install scripts found.
  • 0Remote-code patternsFor example eval(fetch(…)).
  • 396Agent files checked0 phrases flagged for review.
  • 0curl | sh installersIn docs or program code. Information only.
Why this label?

Why this label:

  • 3 critical-severity known vulnerabilities in runtime dependencies (OSV)

Known vulnerabilities in runtime dependencies (OSV): 3 critical, 10 high, 13 medium, 4 low

  • litellm@1.83.0 (PYSEC-2026-3861) in uv.lock
  • litellm@1.83.0 (PYSEC-2026-388) in uv.lock
  • litellm@1.83.0 (PYSEC-2026-391) in uv.lock

Also found (informational, does not change the label):

  • 10 high-severity known vulnerabilities in runtime dependencies (OSV)
  • Skill scripts contact: downloads.rclone.org, docs.blaxel.ai, storage.googleapis.com, api.github.com, github.com, packages.cloud.google.com, s3.

Notes:

  • 3 affected packages only in dev/test/docs manifests, not counted
  • Telemetry/analytics references in code: telemetry

Automated scan of commit e355d69 on 2 Nov 2026, not a manual audit. 1,655 files checked. What we check

Repo activity

From the GitHub API, fetched 2 Nov 2026.

  • 10+Releases, 90 daysLatest v0.24.0, 15 Oct
  • 4dMedian issue closeOver 8 closed issues
  • 30.4kStars+160 since 26 Oct
  • 5kForks33 open issues
  • 2025Repo created11 Mar 2025
30 days ago2 Nov
1 release in the last 30 days, one mark each (GitHub releases feed, last 10 entries).

1 release in the last 30 days (GitHub releases feed, last 10 entries):

Questions about OpenAI Agents SDK (Python)

Answers use this week's StackGrade data.

Is OpenAI Agents SDK (Python) safe to use?

Caution. StackGrade's weekly scan of openai/openai-agents-python gives it a Caution label for 3 critical-severity known vulnerabilities in runtime dependencies (OSV). It does list 10 high-severity advisories in runtime dependencies, which don't change the label. Critical advisories are counted once per package; a known issue in a locked package version does not prove OpenAI Agents SDK (Python) is exploitable in your setup. Automated scan of commit e355d69 on 2 Nov 2026, not a manual audit.

How does StackGrade score OpenAI Agents SDK (Python)?

OpenAI Agents SDK (Python) scores 39 out of 100 in Week 45: #42 of 52 tracked tools overall and #12 in Automation & agent frameworks. The score combines maintenance (32), adoption (25) and docs (69) as percentiles among all 52 tracked tools; community (20% weight) is not measured yet, so the score uses the other 80%. It is not a hands-on test and it is separate from the safety label.

What are the alternatives to OpenAI Agents SDK (Python)?

The highest-scored other tools in Automation & agent frameworks this week: n8n (70, Caution), AutoGPT (62, Caution), CrewAI (57, Caution), Dify (56, Caution). StackGrade has not tested them hands-on.

What licence does OpenAI Agents SDK (Python) use?

The licence listed for openai/openai-agents-python is MIT. Read the licence text before commercial use.

Compare and use it

All facts and sources
Type
Agent framework
License
MIT
Language
Python
Created
11 Mar 2025
Owner
openai (Organization)
Forks
4,970

GitHub description: “A lightweight, powerful framework for multi-agent workflows”

Sources: github.com/openai/openai-agents-python, api.github.com/repos/openai/openai-agents-python, raw.githubusercontent.com/openai/openai-agents-python/main/R. Fetched 2 Nov 2026.

Know when OpenAI Agents SDK (Python) changes label

One email a week: the new top 10, the biggest score changes and every safety label that changed.

Free. Double opt-in: you confirm by email first, unsubscribe in one click.