GraphCanon updated 3w · GitHub synced 3w
Decision brief
WebCanvas provides an open-source framework for building, training, and evaluating web agents with real-time interactions.
Good fit when
- When you need to develop web agents capable of interacting in dynamic online environments rather than static or offline ones.
- For integrating a JavaScript event listener-based evaluation system that decouples assessment from the action space.
Avoid when
- If you are working within fully reproducible environments where Internet dynamics do not influence your project's outcomes.
- When your evaluation needs cannot be addressed by a purely visually-grounded assessment, as WebCanvas decouples its methods from the action space.
- Requirements:
- Requires Python 3.11 for compatibility.
Observed Jul 17, 2026 · Source: enrich:decision_facts
Verify the decision
Maintenance and security
Full trust report- Maintenance
- Dormant (386d since push)
- As of 3w
- Provenance
- Not a fork · Organization account
- As of 3w
- Security (OSV)
- 54 low (54 low)
- As of 1mo
Public GitHub metadata and optional OSV scans. Signals, not a guarantee. Trust methodology.
Install
pip install WebCanvasSimilar tools
Same-category neighbours. No typed graph edges are catalogued for this tool yet.
Evidence and technical details
Sourced facts, taxonomy, compatibility claims, README excerpt, and machine-readable endpoints.
Overview
WebCanvas is an open-source framework to build, train and evaluate web agents.
Capability facts
- Languages
- python
Source: github.language · Jul 28, 2026
Categories
Compatibility
Sourced claims from the README excerpt - not unsourced marketing copy.
Source: README excerpt (regex_v1, Jul 28, 2026)
<a href="https://www.python.org/downloads/release/python-3110/"><img src="https://img.shields.io/badge/pythSource link
Tags
README
WebCanvas: All-in-one Open-Source Framework for Building, Training, and Evaluating Web Agents
Platform •
Dataset •
Discord •
Existing frameworks for web agent development are either offline and static, or operate within a fully reproducible environment with limited Internet dynamics. The WebCanvas project aims to pioneer the online development, training and evaluation of web agents. We offer a suite of toolkits for scaling and maintaining a KEY-NODE based web trajectories for web agents to support this endeavor. We welcome any constructive feedback on the project and look forward to partnering with you in developing agents for web tasks!
🔥 News
- [2024, December 26] v0.0.4 released! Major updates include:
- Introduced a new JavaScript event listener-based evaluation system that decouples evaluation methods from action space, enabling assessment of purely visually-grounded agents
- Integrated with Browserbase cloud browser environment for more stable and consistent evaluation
- Published and maintaining an up-to-date leaderboard for Mind2Web-Live benchmark, still far from saturation!
- [2024, September 9] Support evaluation for OpenAI new o1 models, includes o1-preview and o1-mini. Just set the 'planning_text_model' parameter to 'o1-preview' or 'o1-mini'.
- [2024, August 9] We're excited to announce the release of v0.0.3 of WebCanvas! This update introduces support for evaluation of data operations, such as caching data in process and outputting the final answer. You can now define and evaluate a broader range of web tasks using iMean Builder and WebCanvas. Additionally, we've introduced a new metric: US dollar consumption / key node completion(usd_efficiency_score), detailed in [this section]
For agents
This page has a .md twin and JSON over the API.