How to get picked for agent frameworks by coding agents
Vercel AI SDK took 14% of 681 judged agent frameworks sessions. What the numbers say a vendor in this category should do.
If you sell agent frameworks, this page is the part of the market no dashboard shows you: what a coding agent does when a developer asks for agent frameworks and never compares vendors.
The numbers come from 681 judged sessions with Claude Code, Codex and Cursor, spread across 13 realistic codebases, with every session read by a judge.
What coding agents choose for agent frameworks
Across 681 judged sessions, Vercel AI SDK was chosen most often, in 14% of runs. Cursor SDK was second with 14%. In 25% of runs the agent wrote the code itself and adopted no product at all.
| # | Product | Runs won | Share |
|---|---|---|---|
| 1 | Built in-house (no product adopted) | 170 | 25% |
| 2 | Vercel AI SDK | 95 | 14% |
| 3 | Cursor SDK | 94 | 14% |
| 4 | Inngest | 50 | 7% |
| 5 | Temporal | 29 | 4% |
| 6 | Prism | 28 | 4% |
| 7 | OpenAI Agents SDK | 24 | 4% |
| 8 | LangGraph | 23 | 3% |
| 9 | Azure AI Foundry Agent Service | 19 | 3% |
| 10 | Claude Agent SDK | 16 | 2% |
Full board, every run replayable: the agent frameworks leaderboard.
What the shape of this category means
The leader takes only 14% of runs. This category is genuinely open and the ordering can be moved.
With the top product at 14%, agent frameworks is decided in the moment, from what the agent reads and what it finds in the repository. Nothing is locked in, which is the best situation a vendor can be in and the one where the work pays fastest.
The order here is set by the quality of what an agent can read and by whether your product is already present in the codebase. Both are things you can change.
The agents do not agree with each other
In this category the three agents we ran put different products first.
| Agent | Runs | Picked most often |
|---|---|---|
| Claude Code | 237 | Vercel AI SDK (34) |
| Codex | 243 | Vercel AI SDK (39) |
| Cursor | 201 | Cursor SDK (94) |
Even where they agree, they get there differently. Codex ran a web search in 53% of decision runs and Claude Code in 1.6%, so what you publish reaches one of them in half its agent frameworks sessions and the other in almost none.
Who is asking changes the answer
Every request in this experiment was written as a specific kind of person. In this category the leader changes with the person.
| Who is asking | Runs | Picked most often |
|---|---|---|
| Vibe coder | 163 | Vercel AI SDK |
| Junior developer | 157 | Prism |
| Senior engineer | 254 | Inngest |
| Enterprise team | 107 | Cursor SDK |
That is 4 different products winning agent frameworks for 4 kinds of buyer, out of the same 681 sessions. Nobody here is winning agent frameworks. They are each winning one kind of buyer.
If you sell to more than one of them, you need pages for each. See how to win the enterprise persona.
What you are really competing against
In 25% of runs, the agent wrote the code itself. That makes hand-written code the largest single competitor in this category, larger than most vendors in the table above.
This changes the job of your content. Before you argue that you are better than another vendor, you have to argue that the problem is harder than it looks. What breaks at volume. Which edge cases cost a weekend. What the maintenance actually costs after six months.
That argument has to exist as a page an agent can read, with specifics and numbers. "It is harder than you think" is not an argument. "Here are the four failure modes and what each one costs" is.
Considered, and never chosen
Because the judge records every product an agent raised and not only the one it picked, this board also shows who kept reaching the shortlist and losing. In agent frameworks the clearest case is OpenAI Assistants: on the table in 70 sessions, chosen in none.
| Product | Raised in | Chosen in |
|---|---|---|
| OpenAI Assistants | 70 sessions | 0 |
| LlamaIndex | 64 sessions | 0 |
| Restate | 60 sessions | 0 |
| Semantic Kernel | 31 sessions | 0 |
| CrewAI | 28 sessions | 0 |
Being rejected is a better position than being unknown, and a cheaper one to fix. The product is already in the agent's head and on the list. Whatever ended those 253 sessions is recorded in each transcript, one reason at a time.
What to do about it in agent frameworks
- Argue that the problem is harder than it looks, before you argue that you are better than a rival. 25% of agent frameworks sessions ended in hand-written code, so in roughly one session in 4 no vendor was in the running at all. Write the failure modes and the year-two maintenance cost, with numbers, in the documentation rather than the blog.
- Treat the ordering as movable. The leader holds 14%, so agents are deliberating rather than defaulting, and the inputs they use can change the answer. This is the most winnable shape a category comes in.
- Pick which buyer you are for. The same agent frameworks need written as a vibe coder landed on Vercel AI SDK, and written as an enterprise team landed on Cursor SDK. Those are two markets, and the enterprise one needs pages containing the constraint words: audit log, data residency, retention, single sign-on. See how to win the enterprise persona.
- Measure per agent. Claude Code put Vercel AI SDK first, Codex put Vercel AI SDK first, Cursor put Cursor SDK first. A blended number for agent frameworks describes a market that does not exist.
The work that applies to every category rather than to this one is written up separately: audit your documentation, write a quickstart an agent can follow, and how to measure install share.
Every agent framework on this board
One page per product, with its install share, the per-agent split, and how often it was raised without being chosen.
- Do coding agents recommend Vercel AI SDK? — chosen in 14% of sessions
- Do coding agents recommend Cursor SDK? — chosen in 14% of sessions
- Do coding agents recommend Inngest? — chosen in 7% of sessions
- Do coding agents recommend Temporal? — chosen in 4% of sessions
- Do coding agents recommend Prism? — chosen in 4% of sessions
- Do coding agents recommend OpenAI Agents SDK? — chosen in 4% of sessions
- Do coding agents recommend LangGraph? — chosen in 3% of sessions
- Do coding agents recommend Azure AI Foundry Agent Service? — chosen in 3% of sessions
- Do coding agents recommend Claude Agent SDK? — chosen in 2% of sessions
- Do coding agents recommend DBOS Transact? — chosen in 2% of sessions
- Do coding agents recommend Claude Managed Agents? — chosen in 2% of sessions
- Do coding agents recommend Spring AI? — chosen in 2% of sessions
- Do coding agents recommend Laravel AI SDK? — chosen in 1% of sessions
- Do coding agents recommend Anthropic SDK? — chosen in 1% of sessions
- Do coding agents recommend OpenAI Assistants? — raised in 70 sessions, chosen in none
- Do coding agents recommend LlamaIndex? — raised in 64 sessions, chosen in none
- Do coding agents recommend Restate? — raised in 60 sessions, chosen in none
Where these numbers come from
681 judged sessions in agent frameworks across 13 codebases, part of a published set of 5,292. Real coding agents at pinned versions, in sandboxes, inside realistic codebases, with a simulated project owner in the loop and a blind judge on every session. The full method is on one page: how we measured this.
Every agent frameworks run can be replayed on the board.
<!-- generated by scripts/write-data-pages.mjs -->
Common questions
How many codebases is this based on?
681 judged sessions across 13 realistic codebases. A category only runs on repositories where its seam is open, so coverage differs: some categories ran on more than ten codebases and some on two.
What agent framework do coding agents choose?
Across 681 judged sessions, Vercel AI SDK was chosen most often, in 14% of runs. Cursor SDK was second with 14%. The result changes by agent and by who is asking.
Do Claude Code and Codex pick the same agent framework?
No. Claude Code picked Vercel AI SDK, Codex picked Vercel AI SDK, Cursor picked Cursor SDK. Measuring one agent tells you about part of the market only.
How often do agents build agent frameworks themselves instead of installing something?
In 25% of runs the agent wrote the code itself rather than adopting a product. That makes hand-written code one of the strongest competitors in the category.
How can a vendor improve its position here?
Make the quickstart run when pasted, state the current version on the documentation page, use one name across product, package and import, write pages for the symptoms users describe rather than only the category name, and get into the repository through templates and framework integrations.
Which agent frameworks do agents consider but never choose?
OpenAI Assistants (raised in 70 sessions, chosen in none), LlamaIndex (raised in 64 sessions, chosen in none), Restate (raised in 60 sessions, chosen in none), Semantic Kernel (raised in 31 sessions, chosen in none), CrewAI (raised in 28 sessions, chosen in none). Being considered and not chosen is a different problem from being unknown, and it is usually fixable.
Where this comes from
Armature ran 5,292 judged sessions with Claude Code, Codex and Cursor inside 51 realistic codebases, and published every run. The numbers on this page come from that work.