CAPABILITY · REPLAY

Compile a task once, replay it forever with zero model calls

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: 0
Capability

What deterministic replay does

When a planner solves a task, Twin captures the exact sequence of actions and turns it into a skill. Replaying that skill is deterministic execution — no planning, no model in the loop, no per-run token bill. The same inputs produce the same steps, which is what makes the work cheap and auditable.

Zero LLM calls on replay

A compiled skill runs as a program, not a prompt. Replays cost a flat credit and never touch a model.

Parameterized, not brittle

Skills take typed inputs, so one compiled flow handles every customer, date, or record — not just the example it was trained on.

Same steps, every time

Deterministic execution means a flow that passed in testing behaves identically in production, run after run.

Self-healing handoff

If a page genuinely changed, the replay can fall back to a single re-plan and recompile — so drift fixes itself instead of failing silently.

How it works

From a goal to deterministic action

  1. 1Solve onceThe planner completes the goal on a live page and records every action it took.
  2. 2Compile to a skillThe successful trace is parameterized and stored as a replayable skill with a stable id and version.
  3. 3Replay deterministicallyLater runs execute the skill directly against the DOM — no model, flat credit, identical steps.
  4. 4Recompile on real driftIf the target page changed enough to break a step, Twin re-plans that step once and bumps the skill version.
In practice

See it on a real call

Running a compiled skill is deterministic — four steps, zero model calls, one flat credit.

replay.shbash
curl https://api.twin-browser.com/v1/skills/book-slot/run \
  -H "Authorization: Bearer $TWIN_KEY" \
  -d '{ "inputs": { "day": "Tuesday", "slot": "09:00" } }'

# < x-twin-skill: book-slot@v3
# < x-twin-llm-calls: 0
# < x-twin-steps: 4
api.twin-browser.com
  1. Solve oncedone
  2. Compile to a skillrunning
  3. Replay deterministicallyqueued
  4. Recompile on real driftqueued
At a glance

What deterministic replay is

The facts — how it works, what it costs, and the signal you get back on every call.

PropertyTwin Browser
ExecutionDeterministic, no model in loop
InputsTyped, parameterized
VersioningStable skill id + version
Replay costFlat credit per run
Drift handlingSingle re-plan + recompile
AuditStep list + session video
FAQ

Deterministic replay — common questions

What is the difference between a run and a skill?
A run is one execution of a goal. When a run succeeds, Twin compiles it into a skill — a reusable, parameterized program you can replay deterministically without the planner.
What if the website changes?
Deterministic replay detects when a step no longer matches the page, re-plans just that step once, and recompiles a new skill version. Stable pages never pay that cost.
Is replay really zero model cost?
Yes — replay executes the recorded program against the DOM. The only model spend is the original compile (and any recompile on real drift).

Put your agent to work — you stay in control.

Start free. Hand your agent a goal on any site you authorize, set the guardrails, and let it do the work — repeated runs compile into skills that replay at near-zero cost.