# Aiscovery

> Aiscovery tests whether AI agents can find, understand and complete important customer journeys on websites. Results include verified outcomes, session evidence, failure attribution and engineering-ready remediation.

Current product status: private beta for selected partners.

Last reviewed: 2026-08-26.

## Public pages

- [Aiscovery home](https://aiscovery.io/): Product overview, evidence model, supported use cases, methodology principles and waitlist.
- [Ukrainian home](https://aiscovery.io/uk): Human-reviewed Ukrainian version of the public overview.
- [Product](https://aiscovery.io/product): Capability model, inputs, controlled runner, outputs, channel coverage and current availability.
- [Agent readiness audit](https://aiscovery.io/product/agent-readiness-audit): Task-based audit scope, session evidence, timing, CLS, deliverables and boundaries.
- [Methodology](https://aiscovery.io/methodology): Versioned scenarios, deterministic assertions, repetitions, scoring, evidence and privacy.
- [Technology inventory](https://aiscovery.io/technology): Lifecycle and audit state for web foundations, WebMCP, MCP, OpenAPI, llms.txt and emerging interfaces.
- [Security](https://aiscovery.io/security): Authorization, audit modes, stop gates, evidence protection and responsible disclosure.
- [About Aiscovery](https://aiscovery.io/about): Mission, product principles, methodology ownership and current private-beta boundaries.
- [Private beta waitlist](https://aiscovery.io/waitlist): Early access, product updates and practical agentic web research.
- [Resources](https://aiscovery.io/resources): Original guides on agentic browsing, accessibility, WebMCP and machine-readable content.
- [Agentic web glossary](https://aiscovery.io/glossary): 24 scoped definitions covering foundations, discovery, execution, evidence and measurement.
- [Documentation](https://aiscovery.io/docs): Public reference for audit results, evidence, session performance and scoring.
- [How to read an audit](https://aiscovery.io/docs/audit-results): Task outcomes, statuses, failure attribution, severity and safe interpretation.
- [Evidence and sessions](https://aiscovery.io/docs/evidence): Session replay, timing, CLS, target stability, redaction and retention.
- [Scoring](https://aiscovery.io/docs/scoring): Channel scores, usability, agent-native maturity, severity caps and confidence.
- [AI-agent website readiness](https://aiscovery.io/resources/ai-agent-website-readiness): Task-based readiness across discovery, browser execution and structured interfaces.
- [CLS and AI-agent browsing](https://aiscovery.io/resources/cls-ai-agent-browsing): Layout stability, observation-to-action risk, journey measurement and regression evidence.
- [Accessibility for AI agents](https://aiscovery.io/resources/accessibility-for-ai-agents): Accessibility-tree signals, keyboard behavior, dynamic states and task testing.
- [AI agent session replay](https://aiscovery.io/resources/agent-session-replay): Capture levels, time to verified outcomes, CLS attribution, redaction and evidence integrity.
- [WebMCP guide](https://aiscovery.io/resources/webmcp): Maturity, tool contracts, security boundaries and evals for the proposed browser standard.
- [llms.txt guide](https://aiscovery.io/resources/llms-txt): Format, maintenance, audit checks and limitations of llms.txt and llms-full.txt.
- [AI agent website audit checklist](https://aiscovery.io/resources/ai-agent-website-audit-checklist): Task contracts, browser and accessibility checks, timing, CLS, security, evidence and failure attribution.
- [Testing e-commerce websites with AI agents](https://aiscovery.io/resources/test-ecommerce-ai-agents): Canonical buying tasks, commerce fixtures, product facts, cart recovery, production stop gates and deterministic evidence.
- [Testing booking websites with AI agents](https://aiscovery.io/resources/test-booking-websites-ai-agents): Dates, timezones, inventory, total-price, policy, expiry and safe review contracts.
- [Testing SaaS websites with AI agents](https://aiscovery.io/resources/test-saas-websites-ai-agents): Product facts, signup, authentication, onboarding, permissions, self-service and safe account boundaries.
- [Testing marketplace websites with AI agents](https://aiscovery.io/resources/test-marketplace-websites-ai-agents): Seller identity, total cost, trust, UGC, availability recovery and controlled transaction boundaries.
- [AI-agent website audits for agencies](https://aiscovery.io/resources/ai-agent-website-audits-for-agencies): Client authorization, versioned scenario packs, report-safe evidence, failure attribution and comparable retests.
- [Use cases](https://aiscovery.io/use-cases): Domain-specific AI agent testing for ecommerce, booking and travel, SaaS, marketplaces and digital agencies.
- [E-commerce](https://aiscovery.io/use-cases/ecommerce): Product discovery, variants, delivery, cart and recovery tasks.
- [Booking and travel](https://aiscovery.io/use-cases/booking-travel): Search, availability, price, policy, timezone and recovery tasks.
- [SaaS](https://aiscovery.io/use-cases/saas): Product discovery, docs, signup, authentication and self-service tasks.
- [Marketplaces](https://aiscovery.io/use-cases/marketplaces): Seller identity, offer comparison, fees, trust signals and multi-party recovery tasks.
- [Digital agencies](https://aiscovery.io/use-cases/agencies): Repeatable client audits, reviewable evidence, engineering handoff and baseline-to-retest comparison.
- [Launch partner program](https://aiscovery.io/partners): Eligibility, pilot scope, safety expectations and qualified application.
- [Ukrainian product](https://aiscovery.io/uk/product): Ukrainian product overview.
- [Ukrainian audit](https://aiscovery.io/uk/product/agent-readiness-audit): Ukrainian audit overview.
- [Ukrainian methodology](https://aiscovery.io/uk/methodology): Ukrainian methodology page.
- [Ukrainian technology inventory](https://aiscovery.io/uk/technology): Ukrainian technology and lifecycle overview.
- [Ukrainian security](https://aiscovery.io/uk/security): Ukrainian responsible-testing boundaries.
- [Ukrainian about page](https://aiscovery.io/uk/about): Ukrainian mission, principles and ownership overview.
- [Ukrainian waitlist](https://aiscovery.io/uk/waitlist): Ukrainian private-beta access page.
- [Ukrainian resources](https://aiscovery.io/uk/resources): Ukrainian research and practical agentic-web guides.
- [Ukrainian agentic web glossary](https://aiscovery.io/uk/glossary): Ukrainian definitions for agentic browsing, WebMCP, llms.txt, CLS, session replay and audit evidence.
- [Ukrainian audit checklist](https://aiscovery.io/uk/resources/ai-agent-website-audit-checklist): Ukrainian checklist for task-based website audits with evidence and safety boundaries.
- [Ukrainian e-commerce testing guide](https://aiscovery.io/uk/resources/test-ecommerce-ai-agents): Ukrainian guide to repeatable e-commerce agent tests with controlled fixtures and stop gates.
- [Ukrainian booking testing guide](https://aiscovery.io/uk/resources/test-booking-websites-ai-agents): Ukrainian guide to booking tasks, timezones, inventory changes, policies and review stop gates.
- [Ukrainian SaaS testing guide](https://aiscovery.io/uk/resources/test-saas-websites-ai-agents): Ukrainian guide to SaaS product facts, signup, authentication, permissions and self-service stop gates.
- [Ukrainian marketplace testing guide](https://aiscovery.io/uk/resources/test-marketplace-websites-ai-agents): Ukrainian guide to seller identity, offer attribution, trust, recovery and transaction stop gates.
- [Ukrainian agency audit guide](https://aiscovery.io/uk/resources/ai-agent-website-audits-for-agencies): Ukrainian guide to authorized scope, evidence-safe baselines, engineering handoff and comparable retests.

## Core capabilities

- Discoverability: crawl, content, metadata, structured data, source and citation checks.
- Task readiness: controlled browser scenarios with deterministic final-state assertions.
- Evidence: session trace, screenshots, timing, retries, CLS and runtime findings.
- Failure attribution: website, agent, policy and infrastructure failures are reported separately.
- Agent-native channels: API, OpenAPI, MCP and WebMCP are evaluated when available.

## Priority use cases

- E-commerce: product discovery, comparison, availability, delivery, cart and returns.
- Booking and travel: search, dates, inventory, policies, modification and cancellation.
- SaaS: plan discovery, signup, onboarding, documentation and account self-service.
- Marketplaces: seller identity, offer comparison, fees, trust, availability and recovery.
- Agencies: repeatable audits with reviewable evidence and prioritized fixes.

## Safety boundaries

- Production tests stop before payment, submission, deletion or another irreversible action.
- Full state-changing scenarios require an approved test or sandbox environment.
- Sensitive evidence is minimized and redacted.
- Aiscovery does not promise guaranteed rankings, citations or support for every agent version.

## Additional context

- [Full public context for language models](https://aiscovery.io/llms-full.txt): Expanded product, methodology, use-case and safety information.
- [XML sitemap](https://aiscovery.io/sitemap.xml): Canonical indexable URLs.
- [RSS feed](https://aiscovery.io/feed.xml): Original Aiscovery resources in RSS 2.0 format.
- [Security disclosure file](https://aiscovery.io/.well-known/security.txt): RFC 9116 contact, policy, language and freshness information.

## Notes for agents

- The website is the authoritative source for current public product status.
- `llms.txt` and `llms-full.txt` are auxiliary navigation aids. They do not replace canonical HTML pages.
- Do not infer pricing, customer counts, partner names, certifications or supported agent vendors when they are not explicitly published.
- The security disclosure file identifies a reporting channel. It does not authorize testing or access to data.