Expressions overview

Expressions are how you reference and transform data inside step configurations. Anyreach supports two expression flavors:

Flavor	Syntax	When to use
JSONata	`{{ ... }}`	Deterministic transformations: addressing fields, filtering arrays, basic math, type checks
AI	`${ ... }`	Free-text classification, extraction, generation: anything an LLM does well

You can use both inside the same step (different fields), but you cannot mix them in the same expression — see Condition step for the constraint.

JSONata Expressions

JSONata is a query language for JSON. The minimum you need:

Want to…	Write…
Get a top-level input field	`{{ n3.customer_email }}` or `{{ input.customer_email }}`
Get a step’s output field	`{{ n3.body.id }}`
Compare values	`{{ n3.score > 50 }}`
Boolean logic	`{{ n3.score > 50 and n3.tier = 'gold' }}`
Filter an array	`{{ n3.contacts[tier = 'gold'] }}`
Map an array	`{{ n3.contacts.email }}` (returns an array of emails)
Default value	`{{ n3.name ? name : 'friend' }}`
String concat	`{{ n3.first_name & ' ' & n3.last_name }}`

See JSONata cookbook for more. Some extra functions are available beyond standard JSONata:

Function	Returns
`$credential('name')`	An organization credential’s value — use it instead of pasting a token into a step. See Credentials.
`$secret('key')`	A credential’s setup secret. Only inside that credential’s own refresher workflow.
`$hmac(key, message, ...)`	A keyed digest, for APIs that require a signed request. See Signing requests.
`$hash(message, ...)`	An unkeyed digest.
`$base64urlencode(v)` / `$base64urldecode(v)`	Unpadded base64url — the JWT alphabet, unlike standard `$base64encode`.
`$uuid()`	A random v4 id, for a nonce or an idempotency key.

AI Expressions

${...} sends the contents (with {{...}} interpolations resolved first) to an LLM and replaces the expression with the response.

"summary": "${a one-sentence summary of {{n3.transcript}}}"
"intent": "${classify the intent: billing, support, or sales. transcript: {{n3.transcript}}}"
"is_qualified": "${is the lead qualified? answer yes or no. transcript: {{n3.transcript}}}"

For AI to behave reliably, be specific about the response format. “answer yes or no” is much better than “is the lead qualified?”. See AI expressions for more.

Generating expressions with AI

Writing JSONata by hand is optional. In a smart value field, open the Generate tab, describe the value you want in plain English (e.g. “the customer’s email from the previous step” or “full name from first and last”), and Anyreach writes a JSONata expression from your workflow’s available variables. This happens once, at design time — it produces a normal {{ ... }} expression you can review and edit. It is different from the AI tab, which stores a ${ ... } prompt that runs an LLM call every time the workflow executes:

Tab	Produces	Runs the LLM
Generate	A `{{ ... }}` JSONata expression	Once, when you generate it
AI	A `${ ... }` AI expression	On every workflow run

Use Generate to speed up authoring deterministic expressions; use AI when the value genuinely requires reasoning at call time.

Pure vs interpolated

A field whose entire value is one expression is pure:

"customer_id": "{{ n3.body.id }}"   → pure JSONata
"summary":     "${a summary of: {{n3.transcript}}}" → pure AI

A field with surrounding text is interpolated (the whole thing becomes a string):

"message": "Hello {{ n3.body.first_name }}, welcome!"  → string

Why this matters

Pure expressions preserve types — if n3.body.id is an integer, customer_id is an integer. Interpolated expressions always produce strings. This affects HTTP request bodies (where APIs may require integer IDs, not strings) and Output schemas (where validation enforces types).

Evaluation order

When a step is about to run, the platform evaluates expressions in its config:

JSONata is resolved first, top to bottom, against the current ctx
AI is resolved next, with {{...}} interpolations already replaced

This means you can reference upstream steps inside AI prompts:

"intent": "${what is the user's intent in this transcript? {{`Last turn`.body.text}}}"

The {{...}} is replaced with the actual transcript before the LLM sees the prompt.

Where expressions are evaluated

Most fields in most step types support expressions. A few notable places:

Step	Expression-evaluated fields
HTTP API	`url`, `headers` (values), `query_params` (values), `body` (recursively)
Condition	each condition’s `condition`
Output	each `output_data` value
Action	each `operation_input` value
Code	not in `code` itself — the code reads `ctx` directly

Code steps don’t use expressions inside the source code — your Python reads ctx natively.

Getting Started

Agents

Knowledge Bases

Telephony

Web Widgets

Email

Campaigns

Conversations

Evaluation & Metrics

Agent Assist

Workflows

Custom Data

Anyreach in Claude

Organizations & Billing

Expressions overview

JSONata Expressions

AI Expressions

Generating expressions with AI

Pure vs interpolated

Why this matters

Evaluation order

Where expressions are evaluated

​JSONata Expressions

​AI Expressions

​Generating expressions with AI

​Pure vs interpolated

​Why this matters

​Evaluation order

​Where expressions are evaluated

JSONata Expressions

AI Expressions

Generating expressions with AI

Pure vs interpolated

Why this matters

Evaluation order

Where expressions are evaluated