This page defines the key terms used throughout the OpenScouter platform and documentation. If you encounter an unfamiliar term, this is the place to start.
Core Concepts
| Term | Definition | Also Known As |
|---|
| Study | A commissioned accessibility evaluation of a website or app. A study includes multiple testers, each completing their own test. | ”job” in codebase |
| Test | A single tester’s evaluation within a study. Each tester produces one test, which becomes the basis for their individual report. | ”session” in codebase |
| Barrier | An accessibility obstacle identified during testing. Barriers are documented with severity, context, and recommended fixes. | Accessibility issue, finding |
| Accessibility Score | A 0-100 rating of a site’s accessibility, calculated from the barriers found during a study. Higher scores indicate fewer and less severe barriers. | Score |
| Tester | A neurodivergent expert who performs accessibility testing on behalf of a client. Testers bring lived experience alongside structured evaluation skills. | ND tester |
| Client | A business or organisation that commissions an accessibility study. Clients receive a final report summarising findings across all testers. | Organisation |
| ND Category | One of 13 neurodivergent categories used to group testers by their primary neurodivergent identity. Examples include ADHD, Autism, and Dyslexia. | Neurodivergent category |
AI Agents
OpenScouter uses a pipeline of four AI agents to process test data and generate reports. Each agent has a distinct role.
| Term | Definition | Also Known As |
|---|
| Scouty | The first AI agent in the pipeline. Scouty performs an initial analysis of raw test data, identifying patterns and flagging potential barriers. Also the name of the OpenScouter fox mascot. | Agent 1 |
| Analyst | The second AI agent. The Analyst performs a deep-dive analysis of the data surfaced by Scouty, adding context and severity assessments to each finding. | Agent 2 |
| Report Writer | The third AI agent. The Report Writer generates the final accessibility report for each individual test, translating findings into clear, structured documentation. | Agent 3 |
| Synthesizer | The fourth AI agent. The Synthesizer combines findings from all tests within a study into a single cohesive report for the client. | Agent 4 |
Process Terms
| Term | Definition | Also Known As |
|---|
| Capacity Quiz | An assessment completed before a tester begins a study. It confirms the tester understands their rights and the evaluation process. It is not a test of ability or knowledge. | Onboarding quiz |
| Human-in-the-Loop | The requirement for a tester to review and confirm AI-generated findings before a report is finalised and sent. No report is generated without explicit tester approval. | HITL |
| Plain English | Non-technical summaries of accessibility findings, written so that clients without a technical background can understand and act on the results. | Plain language summary |
| Term | Definition | Also Known As |
|---|
| Calm Mode | A display option that removes decorative visual elements and reduces overall visual noise in the platform interface. Designed for testers who find busy layouts distracting or overwhelming. | Reduced motion mode |
| Scouty | The OpenScouter fox mascot. Also the nickname for Agent 1, which performs the initial analysis of test data. Scouty appears in the testing extension as a friendly companion. | Agent 1, fox mascot |