# What coding agents choose in a Python project > 983 judged sessions inside Python codebases. The leader differed from the overall leader in 4 of 7 categories. Source: https://armature.tech/library/what-coding-agents-choose-in-python Published: 2026-09-03 Publisher: Armature, Inc. (https://armature.tech) --- The single largest factor in which product a coding agent installs is the repository it is working in. Language is the biggest part of that. This page is what the numbers look like inside Python codebases. ## What coding agents choose in Python > Across **983 judged sessions** inside Python codebases, the category leader differed from the overall leader in **4 of the 7 categories** that ran on more than one Python codebase. | Category | Leader in Python | Share here | Leader overall | Share overall | Codebases | | --- | --- | --- | --- | --- | --- | | Evals | Langfuse | 32% | the same | 34% | 1 (see below) | | Agent frameworks | Temporal | 15% | Vercel AI SDK | 14% | 3 | | Databases | Neon | 68% | the same | 66% | 2 | | Code sandboxes | Modal | 51% | E2B | 42% | 2 | | Observability | Amazon CloudWatch | 35% | Sentry | 37% | 2 | | AI gateways | Portkey | 42% | the same | 21% | 3 | | Authentication | WorkOS AuthKit | 42% | the same | 26% | 4 | | Deployment | Render | 87% | Vercel | 41% | 1 (see below) | | Bot protection | django-axes | 31% | Cloudflare Turnstile | 57% | 1 (see below) | | Product analytics | PostHog | 53% | the same | 53% | 1 (see below) | | Cloud | Google Cloud | 100% | AWS | 62% | 1 (see below) | | Performance testing | pytest-benchmark | 42% | JMH | 10% | 1 (see below) | | Serverless compute | AWS Lambda | 60% | the same | 24% | 1 (see below) | | Search | Typesense | 15% | Postgres Full-Text Search | 17% | 1 (see below) | | File storage | Google Cloud Storage | 67% | Amazon S3 | 46% | 2 | Each row is at least 25 judged sessions inside a Python codebase. Categories with fewer are left out rather than reported thin. **Read the single-codebase rows as one repository, not as Python.** 8 of these 15 categories ran on one Python codebase, because the panel only had one with that seam open. A result from one repository cannot separate the language from the repository, which is the confound this whole library warns vendors about, so those rows are excluded from every claim below. They are: evals, deployment, bot protection, product analytics, cloud, performance testing, serverless compute, search. ## The same leader, a different grip In 2 categories the name at the top did not change but its hold did, by twelve points or more. That is still a different market. | Category | Leader | Share in Python | Share overall | Difference | | --- | --- | --- | --- | --- | | AI gateways | Portkey | 42% | 21% | +21 points | | Authentication | WorkOS AuthKit | 42% | 26% | +16 points | A leader that is stronger here has the Python templates and the Python documentation right. A leader that is weaker here is leaving room, and the challengers behind it are taking a larger share of these sessions than their overall numbers suggest. ## Where Python disagrees with the overall result 4 categories. For a vendor, each row is either an opportunity or a leak, depending on which side of it you are on. **Agent frameworks.** In Python codebases the agents chose Temporal most often, in 15% of 132 sessions. Across all languages the leader is Vercel AI SDK at 14%. See [the Agent frameworks playbook](/library/agent-frameworks-coding-agents-playbook). **Code sandboxes.** In Python codebases the agents chose Modal most often, in 51% of 108 sessions. Across all languages the leader is E2B at 42%. See [the Code sandboxes playbook](/library/sandboxes-coding-agents-playbook). **Observability.** In Python codebases the agents chose Amazon CloudWatch most often, in 35% of 72 sessions. Across all languages the leader is Sentry at 37%. See [the Observability playbook](/library/observability-coding-agents-playbook). **File storage.** In Python codebases the agents chose Google Cloud Storage most often, in 67% of 27 sessions. Across all languages the leader is Amazon S3 at 46%. See [the File storage playbook](/library/storage-coding-agents-playbook). ## Where agents write it themselves in Python | Category | Sessions that adopted no product | Codebases | | --- | --- | --- | | Performance testing | 56% | 1 | | Search | 53% | 1 | | Product analytics | 42% | 1 | | Evals | 35% | 1 | | Agent frameworks | 33% | 3 | In those categories your first competitor is a hand-written implementation, not another vendor. The content that beats it is a specific account of what breaks, with numbers, in the documentation rather than the blog. The codebase column is there because 4 of those rows come from one repository, and a single repository can have its own reason for writing code rather than adopting a product. Weigh them accordingly. ## Why the split matters 4 leaders swapped and 2 held their place with a different grip, out of the 7 categories measured across more than one codebase. Nobody changed a product or rewrote a page between those sessions. The mechanism is covered in [the repository decides more than your marketing does](/library/what-decides-the-pick-the-repository). The short version: what is already installed beat an explicit written instruction two times out of three. ## If you sell into Python Split your own install share by language before you read it. A vendor in ai gateways looking at 21% overall is looking at 42% here, across 3 codebases. Then ship an official Python template for the stacks your users run, and check your package name in the Python registry against your product name. The rest of the work is not language-specific: [audit your documentation](/library/audit-your-docs-for-coding-agents) and [write a quickstart an agent can follow](/library/write-a-quickstart-an-agent-can-follow). ## How this was measured The 983 Python sessions are part of a published set of 5,292, run with real coding agents inside realistic codebases and judged blind. Full method: [how we measured this](/library/how-we-measured-this). Every run is on [the sector boards](/leaderboards/sectors). ## Common questions ### What tools do coding agents choose in a Python project? Across 983 judged sessions inside Python codebases, the leader differed from the overall leader in 4 of the 15 categories measured. In Evals, Python projects got Langfuse in 32% of sessions. ### Does the language of the repository change which tool an agent picks? Yes. In Python codebases the winner differed from the overall winner in 4 categories: Agent frameworks (Temporal rather than Vercel AI SDK), Code sandboxes (Modal rather than E2B), Observability (Amazon CloudWatch rather than Sentry), File storage (Google Cloud Storage rather than Amazon S3). ### In which categories do agents write their own code in Python? Performance testing (56% of sessions), Search (53% of sessions), Product analytics (42% of sessions). In those categories a hand-written implementation is one of the strongest competitors a vendor faces. ### How was this measured? Real coding agents at pinned versions ran in sandboxes inside realistic Python codebases with real lock files and replayed git histories. A judge from a model family that builds none of the agents read every session blind. ## Read next - [The repository decides more than your marketing does](https://armature.tech/library/what-decides-the-pick-the-repository) (Markdown: https://armature.tech/library/what-decides-the-pick-the-repository.md) - [Agent discoverability: the complete guide](https://armature.tech/library/agent-discoverability) (Markdown: https://armature.tech/library/agent-discoverability.md) - [How to get picked for evals by coding agents](https://armature.tech/library/evals-coding-agents-playbook) (Markdown: https://armature.tech/library/evals-coding-agents-playbook.md) - [How to get picked for agent frameworks by coding agents](https://armature.tech/library/agent-frameworks-coding-agents-playbook) (Markdown: https://armature.tech/library/agent-frameworks-coding-agents-playbook.md) --- Armature helps software products get discovered and used by coding agents. Service: https://armature.tech/discoverability · Results: https://armature.tech/leaderboards/sectors · Contact: contact@armature.tech