Skip to content
agentboards
Browse sections All boards

About & disclosure

A practical reference for AI developer tools.

AgentBoards helps developers compare coding agents, AI editors, review tools, frameworks and models. Start with your workflow, inspect the evidence, then test a small shortlist on your own work.

What you can check here

  • Source-linked specifications. Tool profiles link to provider documentation, repositories and pricing. A verification date records a source check, not a promise that every detail remains current.
  • Published model benchmarks. The model board retains exact tested configurations, evaluator attribution and snapshot dates. Benchmark task costs are not estimates for your workload.
  • Practical evaluations. The test kit provides repeatable tasks and blank result records. It does not claim that AgentBoards has run competing agents or established a winner.
  • Release updates. The news feed links to release sources. Automated selection rules surface changes; dated editorial briefings are maintained separately.

AI disclosure and current maintenance

Previously published reviews by Los Agentes were generated by AI personas. They are labeled opinions, not human testimonials or evidence of hands-on testing. Their original review dates remain visible, and previously published reviews are retained.

Paid AI generation has been removed from the website. Opening a page, searching, comparing models or tools, and using the test kit do not send prompts to Gemini or another model. Existing Gemini product listings describe tools in the catalog; they do not indicate that AgentBoards is running Gemini.

Listings and editorial changes are maintained from sources without model inference. Non-AI collection can refresh public repository metrics and release information; it does not silently verify every specification or regenerate critic reviews.

How to read the scores

Agent board scores combine historical critic opinions, measured adoption and source coverage. Model benchmarks are a separate evidence source. Human reviews are labeled separately. Read the published methodology before using a rank to make a decision.

Corrections and builder responses

Submit a correction with a source when information is incomplete or outdated. Builders can claim a listing and respond on the record. Sponsorship does not change a ranked score.

Open data

The JSON API exposes the dataset without an account or API key. Data is licensed CC BY 4.0 and code is MIT. Attribute AgentBoards when reusing the data. The API is generated from the published catalog; its response time is not a new verification date for each record.

Loading agent listings… Models and guides are ready.
↑↓ navigate · Enter open · Esc closeModels & guides