1.6K+ listed tools · 150+ categories
1.6K+listed tools
150+categories
Explore
By capability
150+ categoriesBy ranking
Just released
Newest listings and revisions first
Featured
Sponsored placements
Guides
Research
Why it matters: Teams using agents to summarize experiments, code changes or completed tasks can test explicit flaw-disclosure instructions and audit reports against underlying logs. The preprint’s synthetic evaluations offer a test design, rather than a guarantee that a short prompt makes production reports trustworthy.
AI HOT SelectedRead source
Tools
Why it matters: Hardware hobbyists can prototype physical interfaces to an existing AI assistant using the Apache-licensed code. Muse access through the token is restricted to personal, non-commercial use, and the project is unsupported, so it should be evaluated as a hobby integration rather than a supported commercial platform.
TechCrunch AIRead source
Research
Why it matters: Agent builders can use the existing benchmark and OpenEnv adapter to compare models on repeated execution and inspect failures that leave incorrect records. The expanded results help frame model selection, but synthetic-task scores and estimated token costs do not establish reliability or costs in a production workflow.
Hugging Face BlogRead source
How it works
Each list is an order an editor sets for one category. It changes only when a new order is published.
Ranking methodologyWhere we review a tool, we use written standards. It isn't a universal hands-on score.
Selection standardsFeatured spots rotate daily and never show a rank number.
See featured toolsSearch if you already know the tool or task. If you know the capability you need, open its category; if you know your line of work, start from an industry guide; if you want a ready-made shortlist, open a published ranking. Confirm plans and limits on the vendor's official site before you buy — they change.
The selection standards are editorial criteria, not a claim that every listed tool received the same hands-on benchmark. Verified Listings also pass the paid submission review described on the submission page.
Listings show a version and a public starting price where one is known. ToolWorthy does not recheck every listing on a single schedule, so confirm current tiers, credits and usage limits with the vendor.
ToolWorthy does not currently provide a universal interactive side-by-side comparison tool; published decision guides compare selected options with stated source limits. ToolWorthy also does not currently document a public developer API — contact support for a specific data or licensing inquiry.
ToolWorthy records public free or freemium availability where known, but vendors can change access, credits and commercial-use terms. Confirm the current plan before relying on it.
Search the directory
