Sensei
sensei.sh
- Free plan
- API
6.7 Club score Visit sensei.sh
What the maker publishes
- Agent connection
- The site shows agents connecting to Sensei over HTTP or stdio. source
- Built-in suites
- The site lists built-in suites for SDR, customer support, content writing, bartender, dungeon master, and cat interview roles. source
- CI/CD
- The repository lists CI/CD integration among Sensei's capabilities. source
- Community library
- The home page describes a library of 70+ community-built professional evaluation suites. source
- Custom suites
- The site says users can create their own evaluation suites. source
- Custom suites
- Users can create their own evaluation suites and publish them to the community marketplace. source
- Evaluation
- It evaluates agents across task execution, reasoning, and self-improvement. source
- Evaluation layers
- Sensei evaluates task execution, reasoning, and self-improvement through feedback. source
- Integrations
- The repository documents HTTP, stdio, OpenAI-compatible, and LangServe adapters for connecting agents. source
- Intended users
- Sensei is designed for people evaluating AI agents on real-world professional tasks using standardized benchmarks, scenarios, and measurable KPIs. source
- License
- Sensei is distributed under the MIT License. source
- LLM judging
- Sensei supports LLM-based and comparative judging, with OpenAI shown as an example provider. source
- Marketplace
- The marketplace offers community evaluation suites to discover, download, and publish. source
- Marketplace
- The marketplace lets users discover, download, and publish AI-agent evaluation suites. source
- Marketplace categories
- Marketplace categories include engineering, sales, marketing, product, design, support, testing, analytics, compliance, and content. source
- Purpose
- Sensei describes itself as an open-source qualification engine for AI agents. source
- Purpose
- Sensei is an open-source qualification engine for testing, evaluating, and certifying AI agents across professional skills. source
- Reports
- The repository documents machine-readable JSON results and terminal reports. source
- Scoring
- Its evaluation pipeline produces a score from 0 to 100 and a badge decision. source
- Scoring
- Its evaluation pipeline produces a score from 0 to 100 and a badge decision. source
- Security policy
- The security policy supports version 0.x.x and asks vulnerability reporters to contact the maintainer by email rather than opening a public issue. source
- Security response
- The policy says reports are acknowledged within 48 hours, initially assessed within one week, and targeted for a patch within 30 days of confirmation. source
- Test suites
- Built-in suites cover roles including SDR, customer support, content writer, bartender, dungeon master, and cat interview. source
Checked against the maker's pages on 9 Oct 2026.