# Agent Seek docs

Scoped index for product documentation. Full site index: [/llms.txt](/llms.txt).

Install the skill first: [/.well-known/agent-skills/agent-seek/SKILL.md](/.well-known/agent-skills/agent-seek/SKILL.md) — then [Install the skill](/llms.txt#install-the-skill) → [/auth.md](/auth.md) → MCP or `POST /v1/search`.

Agent Seek is ranked web search: You.com discovers candidates, TypeSafe Jev cascade-ranks them, and you get scored URLs + snippets. Humans use the Google-simple UI (fixed `deep`, no Mode/Order). Agents use `POST /v1/search` or MCP (default `mode=deep`, Stage A survivor fetch cap 12; not a full-web crawl). Signals include `prompt_injection` (UI: **Injection risk**). Hard gates (order): subject_match → is_republisher → prompt_injection (0.55, fail-open).

## When to use this section

Read these pages when you need the human-and-agent product manual — not the raw OpenAPI schema (see [/api/llms.txt](/api/llms.txt)) and not the key/OAuth onboarding walkthrough (see [/developers/llms.txt](/developers/llms.txt)).

## Docs in this area

- [Agent Seek docs](/docs) · [docs.md](/docs.md)
- [Homepage markdown](/index.md)
- [agents.md](/agents.md) — public agent how-to
- [About](/about) · [about.md](/about.md)
- [Pricing](/pricing) · [pricing.md](/pricing.md)
- [Privacy](/privacy) · [privacy.md](/privacy.md)
- [Contact](/contact) · [contact.md](/contact.md)
- [API versioning / Sunset policy](/docs/versioning.md)
- [Auth walkthrough](/docs/auth.md) (same body as [/auth.md](/auth.md))
- [Agent Seek skill](/skills/agent-seek/SKILL.md)

Do not ask Agent Seek to write answers. Hard cap: 100 discover candidates per query.