# Web Context Bench (Python)

Local-page benchmark comparing six browser representations on token use, factual reading and target-element selection, with deterministic answer scoring.

- Catalog URL: https://awesomecua.github.io/resources/github-charlypoly-web-context-bench-python/
- Original source: https://github.com/charlypoly/web-context-bench-python
- Kind: benchmark
- Review: curated
- Scope: core
- Environments: web
- Topics: evaluation, grounding, efficiency, screen-understanding
- Evidence checked: 2026-10-20T11:51:30.874397Z
- ID: github:charlypoly/web-context-bench-python

## Limitations

This is a single-snapshot representation evaluation on local fixtures and captured pages, not a live multi-step agent benchmark. Its shared-prompt, one-model setup differs from complete Browser Use or Stagehand agents. The methodology notes unresolved redistribution licensing for third-party practice-site snapshots; MIT is recorded for repository code only. Selected scoring code was read, but no models, browser tasks or claimed results were run or reproduced.

- Record JSON: https://awesomecua.github.io/resources/github-charlypoly-web-context-bench-python/record.json

## Structured profile

- actions: READ text answer, ACT selector, text/label, accessible role/name, DOM index, Stagehand reference or screenshot point, according to the representation
- evaluation: READ uses normalized exact match against an answer or allowed alias. ACT resolves predicted identifiers to ground-truth elements, or checks screenshot points inside the target box. Representation-token accounting subtracts fixed-prompt tokens; screenshots use billed input usage minus the fixed prompt. These are answer/target correctness and token metrics, not completed-task success.
- focus: Browser-representation token use, Factual READ accuracy, ACT target-element selection
- observations: Top-document HTML, Markdown converted from HTML, Playwright accessibility snapshot, Browser Use indexed DOM, Stagehand hybrid snapshot, Viewport screenshot
- setting: offline

### Count observations

No count recorded.

Fixed local-page observations; the reported run uses claude-sonnet-5 and a shared prompt, without scrolling or an agent loop. Browser Use disables its viewport cutoff and sends untruncated text; Stagehand uses a separate matched browser load. Short-string representability labels are approximate. Offline does not mean API-free. Third-party snapshot redistribution remains unresolved; results were not reproduced.

## License declarations

- code: MIT — https://raw.githubusercontent.com/charlypoly/web-context-bench-python/58d604bfaaac60ad42f02013f53f9fb301c1dcda/LICENSE
- data: not recorded
- document: not recorded
- weights: not recorded

- code review: The inspected root LICENSE contains complete MIT terms for repository software. The methodology explicitly leaves redistribution permission for captured third-party practice-site pages unresolved; code MIT is not a blanket snapshot/data license.
  Sources: https://raw.githubusercontent.com/charlypoly/web-context-bench-python/58d604bfaaac60ad42f02013f53f9fb301c1dcda/LICENSE

## Primary evidence

- https://raw.githubusercontent.com/charlypoly/web-context-bench-python/58d604bfaaac60ad42f02013f53f9fb301c1dcda/README.md (retrieved 2026-10-20T11:51:30.874397Z; SHA-256 14775f6e3daba69d4684380fd33bf717bdc3093b65be796eb863bbd27d0c0bab)
- https://api.github.com/search/repositories?q=topic%3Aweb-agents+is%3Apublic+fork%3Afalse+created%3A2008-01-01..2026-10-20&per_page=100&page=1&sort=updated&order=asc (retrieved 2026-10-20T11:05:35.063004+00:00; SHA-256 3e22e1cbfa47bcf98f6f70a5f6f162a31a509a9f2091a84679bb5231977e29d9)
- https://github.com/charlypoly/web-context-bench-python (retrieved 2026-10-20T11:50:54.545491Z; SHA-256 fd30e94e78e7167a49579922259fde1c7d7f8259e9f00afba15faf3a71966a94)
- https://raw.githubusercontent.com/charlypoly/web-context-bench-python/58d604bfaaac60ad42f02013f53f9fb301c1dcda/METHODOLOGY.md (retrieved 2026-10-20T11:52:30.131541Z; SHA-256 1c3c0cbf9f9d9f7cd9e7a2f027ecb217fd5c16fa157916281a15bb7f403f6f1c)
- https://raw.githubusercontent.com/charlypoly/web-context-bench-python/58d604bfaaac60ad42f02013f53f9fb301c1dcda/src/wpbench/scoring.py (retrieved 2026-10-20T11:53:28.567861Z; SHA-256 8cada5d71b5a61e7aff2e1ac2b951c266eb2db7c082c06c5a13aa127d42b3fc5)
- https://raw.githubusercontent.com/charlypoly/web-context-bench-python/58d604bfaaac60ad42f02013f53f9fb301c1dcda/LICENSE (retrieved 2026-10-20T11:51:41.499735Z; SHA-256 9b64ef672f5091b6074d02502f6d3c161d94308a3d9d142d7437dcddf611bbc4)