The platform that makes the next run cheaper than the last
Most browser infra re-runs the model on every execution, so cost climbs with usage. Twin compiles a task once, matches re-phrased requests to it, and replays deterministically with zero model calls — so marginal cost trends toward zero. These are the shipped capabilities behind that wedge.
Semantic dispatch cache
Re-phrased goals fuzzy-match a skill you already compiled — so the second request and every one after it skips the LLM entirely.
POST /api/v1/agents/run → x-twin-cache: hitDeterministic replay
A successful run compiles into a skill — an ordered, parameterized program of browser actions that replays the same way every time.
POST /api/v1/skills/{id}/run → x-twin-llm-calls: 0Cross-tenant skill corpus
Common flows — log in, search, paginate, fill a form — get compiled once and shared through a sanitized corpus, so your agents start ahead.
x-twin-skill-source: corpusToken-efficient DOM state
Twin compiles a live page into a numerically-indexed list of interactive elements under a token budget, so the planner sees signal, not markup.
x-twin-dom-tokens: 2,941Human-in-the-loop handoff
Approvals, MFA on an authorized flow, or an ambiguous choice pause the run, hand off to a person, and continue from the exact same point.
x-twin-status: awaiting_approvalLive view & session video
Stream the browser session in real time, then keep a durable video of every run for debugging, audit, and proof of what happened.
GET /api/v1/runs/{id}/liveAgent & skill library
Every compiled skill is stored, versioned, and reusable — browse them, run them directly, and share them across your agents.
GET /api/v1/skillsCredential vault
Per-tenant encrypted secrets are injected into a run when a step needs them — and never enter the model, the logs, or the skill corpus.
POST /api/v1/secretsProxy & geo support
Route authorized runs through managed residential and datacenter proxies so sites render and behave the way your users actually see them.
x-twin-egress: residential/usBuild on the browser layer for AI agents.
One token-efficient API turns a goal into deterministic, replayable action. Start free and compile your first skill in minutes.