The Looking Glass

What are you trying to do?

207 built-in operators · press / to ask

Every built-in operator is a cascade you can read, extend, or compose. Start with what you’re trying to do, not what it’s called.

Classification

54 operators
dimension

AUDIENCE

Identify target audience via zero-shot NLI

classificationnlispecialist-zootext+1
dimension

AUDIENCE_LLM

LLM-backed audience identification (escape hatch for AUDIENCE)

classificationllmllm-escape-hatchtext
dimension

AUTHENTICITY

Assess authenticity of content via zero-shot NLI

classificationnlispecialist-zootext
dimension

AUTHENTICITY_LLM

LLM-backed authenticity assessment (escape hatch for AUTHENTICITY)

classificationllmllm-escape-hatchtext
dimension

BUCKET

Classify text into user-specified buckets via zero-shot NLI

classificationnlispecialist-zootext
dimension

BUCKET_LLM

LLM-backed bucketing (escape hatch for BUCKET)

classificationllmllm-escape-hatchtext
scalar

CASE

Multi-branch semantic classification via specialist zoo NLI

classificationnlispecialist-zootext
scalar

CASE_LLM

LLM-backed multi-branch semantic classification

classificationllmllm-escape-hatchtext
dimension

CATEGORY

Categorize text via zero-shot NLI with default or custom labels

classificationnlispecialist-zootext
dimension

CATEGORY_LLM

LLM-backed categorization (escape hatch for CATEGORY)

classificationllmllm-escape-hatchtext
scalar

CLASSIFY

Classify single text into one of the provided topics (zero-shot NLI)

classificationnlispecialist-zoojson
aggregate

CLASSIFY_COLLECTION

Classify a text collection into one category (embedding majority vote)

classificationllmscales-largejson
aggregate

CLASSIFY_COLLECTION_LLM

LLM-backed collection classification (escape hatch for CLASSIFY)

classificationllmllm-escape-hatchscales-large+1
scalar

CLASSIFY_LLM

LLM-backed single-text classification (escape hatch for CLASSIFY_SINGLE)

classificationllmllm-escape-hatchjson
dimension

COMPLEXITY

Assess complexity level for grouping (zero-shot NLI)

classificationnlispecialist-zootext
dimension

COMPLEXITY_LLM

LLM-backed complexity assessment (escape hatch for COMPLEXITY)

classificationllmllm-escape-hatchtext
dimension

CREDIBILITY

Analyze credibility/reliability of text descriptions (zero-shot NLI)

classificationnlispecialist-zootext
dimension

CREDIBILITY_LLM

LLM-backed credibility assessment (escape hatch for CREDIBILITY)

classificationllmllm-escape-hatchtext
dimension

DOMAIN

Detect subject matter domain for grouping (zero-shot NLI)

classificationnlispecialist-zootext
dimension

DOMAIN_LLM

LLM-backed subject-matter domain detection (escape hatch for DOMAIN)

classificationllmllm-escape-hatchtext
dimension

ENGAGEMENT

Predict engagement type/level via zero-shot NLI

classificationnlispecialist-zootext
dimension

ENGAGEMENT_LLM

LLM-backed engagement-level prediction (escape hatch for ENGAGEMENT)

classificationllmllm-escape-hatchtext
scalar

EVIDENCE_TYPE

Classify evidence type and strength

classificationllmtext
dimension

FORMALITY

Detect formality level for grouping (zero-shot NLI)

classificationnlispecialist-zootext
dimension

FORMALITY_LLM

LLM-backed formality detection (escape hatch for FORMALITY)

classificationllmllm-escape-hatchtext
pipeline

GENERIC_DISCRIMINATOR

Built-in discriminator for CHOOSE stages without explicit BY clause.

classificationllmpipeline-composablejson
scalar

INFER_TYPE

Detect the semantic type of a value (email, phone, date, etc.)

classificationllmtext
dimension

INTENT

Classify the communicative intent of text (zero-shot NLI)

classificationnlispecialist-zootext
dimension

INTENT_LLM

LLM-backed intent classification (escape hatch for INTENT)

classificationllmllm-escape-hatchtext
dimension

LANGUAGE

Detect language for grouping (xlm-roberta-lid)

classificationhybridspecialist-zootext
dimension

LANGUAGE_LLM

LLM-backed language detection (escape hatch for LANGUAGE)

classificationllmllm-escape-hatchtext
scalar

LOOKS_LIKE

Check if value looks like a type (fuzzy)

classificationnlispecialist-zootext
scalar

LOOKS_LIKE_LLM

LLM-backed looks_like (escape hatch for LOOKS_LIKE)

classificationllmllm-escape-hatchtext
scalar

MATCH_TEMPLATE

Evaluate whether a plain-English statement is true or false

classificationllmtext
dimension

NARRATIVE

Identify narrative frame via zero-shot NLI

classificationnlispecialist-zootext
dimension

NARRATIVE_LLM

LLM-backed narrative-frame identification (escape hatch for NARRATIVE)

classificationllmllm-escape-hatchtext
scalar

QUALITY

Assess data quality (0.0-1.0)

classificationllmtext
scalar

QUALITY_LLM

LLM-backed data quality assessment (0.0-1.0)

classificationllmllm-escape-hatchtext
dimension

SENTIMENT

Dimension-shaped sentiment analysis for GROUP BY

classificationhybridspecialist-zootext
dimension

SENTIMENT_LLM

LLM-backed dimensional sentiment analysis (escape hatch for SENTIMENT)

classificationllmllm-escape-hatchtext
dimension

STANCE

Identify stance on a topic via zero-shot NLI (topic required)

classificationnlispecialist-zootext
dimension

STANCE_LLM

LLM-backed stance detection on a topic (escape hatch for STANCE)

classificationllmllm-escape-hatchtext
scalar

SWITCH

Semantic pattern matching via specialist zoo NLI (MATCH/YIELD syntax)

classificationnlispecialist-zootext
scalar

SWITCH_LLM

LLM-backed semantic pattern matching

classificationllmllm-escape-hatchtext
dimension

TIMEFRAME

Extract timeframe dimension via zero-shot NLI

classificationnlispecialist-zoodate
dimension

TIMEFRAME_LLM

LLM-backed timeframe extraction (escape hatch for TIMEFRAME)

classificationllmllm-escape-hatchtext
dimension

TOPICS

Extract topics from text collection and assign each text to a topic

classificationembedding-modelspecialist-zooscales-large+1
dimension

TOPICS_LLM

LLM-only topic extraction + assignment (escape hatch for TOPICS)

classificationllmllm-escape-hatchscales-large+1
dimension

TOXICITY

Assess toxicity level for grouping (toxic-bert)

classificationhybridspecialist-zootext
dimension

TOXICITY_LLM

LLM-backed toxicity assessment (escape hatch for TOXICITY)

classificationllmllm-escape-hatchtext
dimension

VIBES

Extract vibe categories from a text collection and assign each text to one

classificationhybridspecialist-zooscales-large+1
dimension

VIBES_LLM

LLM-backed vibe extraction and assignment (escape hatch for VIBES)

classificationllmllm-escape-hatchscales-large+1
dimension

VIRALITY

Predict viral potential via zero-shot NLI

classificationnlispecialist-zootext
dimension

VIRALITY_LLM

LLM-backed virality prediction (escape hatch for VIRALITY)

classificationllmllm-escape-hatchtext

Extraction

33 operators
scalar

CRAWL_BATCH

Crawl a website and extract structured data from each page (via Firecrawl)

extractionexternal-apiurl
scalar

EXTRACT

Extract specific information from unstructured text (zero-shot NER)

extractionnlispecialist-zootext
scalar

EXTRACT_LLM

LLM-backed extraction (escape hatch for EXTRACTS)

extractionllmllm-escape-hatchtext
scalar

EXTRACT_STRUCTURED

Extract structured fields from text per a user-supplied schema

extractionllmtext
aggregate

MERGE_TIMELINES

Merge multiple timelines into unified chronological sequence

extractionllmjson
scalar

PARSE

Extract information from text using natural-language instructions

extractionllmtext
scalar

PARSE

Parse, validate, or transform patterned strings using plain-English instructions

extractionllmtextpipeline-composable
scalar

PARSE_ADDRESS

Parse US address into JSON with fields: street_number, street_name, unit, city, state, zip, country, formatted

extractionllmtext
scalar

PARSE_ADDRESS_LLM

LLM-backed address parsing (escape hatch for PARSE_ADDRESS)

extractionllmllm-escape-hatchtext
scalar

PARSE_DATE

Parse date from any format into JSON with fields: year, month, day, iso, formatted, day_of_week, is_valid

extractionllmdate
scalar

PARSE_DATE_LLM

LLM-backed date parsing (escape hatch for PARSE_DATE)

extractionllmllm-escape-hatchdate
scalar

PARSE_DOCUMENT

PDF → structured JSON (pypdf + Donut OCR fallback)

extractionllmscales-largeimage
scalar

PARSE_EMAIL

Parse raw email into structured JSON (from, to, cc, date, subject, body)

extractionllmemail
scalar

PARSE_EMAIL_LLM

LLM-backed email parsing (escape hatch for PARSE_EMAIL)

extractionllmllm-escape-hatchemail
scalar

PARSE_NAME

Parse person name into JSON with fields: prefix, first, middle, last, suffix, nickname, formatted

extractionllmtext
scalar

PARSE_NAME_LLM

LLM-backed name parsing (escape hatch for PARSE_NAME)

extractionllmllm-escape-hatchtext
scalar

PARSE_PHONE

Parse phone number into JSON with fields: country_code, area_code, e164, national, international, is_valid, type

extractionllmtext
scalar

PARSE_PHONE_LLM

LLM-backed phone parsing (escape hatch for PARSE_PHONE)

extractionllmllm-escape-hatchtext
scalar

PARSE_TYPED

Extract structured data from text using a predefined type schema

extractionllmtext
scalar

REFRESH_CRAWL

Re-crawl a stored web-table using its saved configuration

extractionexternal-apiurl
scalar

RELATIONS

Extract (head, type, tail) relation triples from text

extractionllmtext
scalar

RICH_TRIPLES

Extract (subject, predicate, object, evidence) knowledge graph triples with source context

extractionllmtext
scalar

RSS

Fetch an RSS/Atom feed and return entries as a table

extractionllmtext
scalar

SCRAPE

Fetch a URL and return its content as clean markdown (via Firecrawl)

extractionexternal-apiurl
scalar

SMART_JSON

Extract a value from JSON using a plain-English path description

extractionllmjsonpipeline-composable
scalar

SMART_SPLIT

Split a value into components without hardcoding a delimiter

extractionllmtext
scalar

SMART_UNPACK

Extract multiple distinct values from a compound/messy field

extractionllmtext
scalar

TIMELINE

Extract chronologically ordered events from text

extractionllmtext
scalar

TRIPLES

Extract (subject, predicate, object) knowledge graph triples from text

extractionllmtext
scalar

UNNEST_SMART

Expand a compound value into separate elements (handles mixed delimiters)

extractionllmtext
scalar

WEB_AGENT

AI web research agent — no starting URL required (via Firecrawl)

extractionexternal-apiurl
scalar

WEB_EXTRACT

Extract structured data from multiple pages using LLM-powered extraction

extractionexternal-apiurl
scalar

WEB_SCRAPE

Extract structured data from a URL as a table (LLM-powered)

extractionexternal-apiurl

Similarity

10 operators

Implication

6 operators

Summarization

8 operators

Ranking

7 operators

Clustering

1 operator

Search

5 operators

Embedding

7 operators

Analysis

15 operators
scalar

AB_TEST

Bayesian A/B test (Beta-Binomial) returning full posterior + recommendation

analysisdeterministicnumeric
scalar

ASSESS_CONFIDENCE

Assess the confidence/quality of a cascade execution result.

analysisllmtext
aggregate

COMPARE

Compare values in a group for similarities, differences, and patterns

analysisllmtext
scalar

COUNTERARGUMENT

Generate the strongest counterargument to a position

analysisllmtext
scalar

DP_COUNT

Differentially-private count (Laplace mechanism, sensitivity=1)

analysisdeterministicjson
scalar

DP_MEAN

Differentially-private mean (Laplace mechanism)

analysisdeterministicjson
scalar

FALLACY

Detect logical fallacies in an argument

analysisllmtext
scalar

FORECAST

Forecast a univariate time series via zero-shot Chronos-Bolt

analysisdeterministicjson
scalar

KAPLAN_MEIER

Kaplan-Meier survival curve — time-to-event estimator with 95% CI

analysisdeterministicjson
scalar

LATEST

Returns the latest output from a cascade cell

analysisdeterministictext
aggregate

SENTIMENT

Aggregate sentiment score (-1.0 to 1.0) for a collection of texts

analysishybridspecialist-zooscales-large+1
scalar

SENTIMENT_SCALAR

Per-row sentiment score (-1.0 to 1.0) for a text value

analysishybridspecialist-zooscales-large+1
scalar

SHORTEST_PATH

Shortest path between two nodes on an ad-hoc edge list (NetworkX)

analysisdeterministicjson
scalar

STEELMAN

Construct the strongest version of an argument

analysisllmtext
scalar

WEAKNESSES

Identify logical weaknesses and gaps in an argument

analysisllmtext

Cleaning

24 operators
scalar

ANONYMIZE

Remove or mask personally identifiable information from text

cleaningllmtext
scalar

AS_SMART

Type-cast messy real-world values that trip up standard CAST

cleaningllmtext
scalar

CANONICAL

Return the canonical/official form of a value (auto-detects entity type)

cleaningllmtext
scalar

CLEAN_YEAR

Extracts 4-digit year from messy text, returns -1 if undetermined

cleaningllmdate
scalar

CLEAN_YEAR_LLM

LLM-backed year extraction (escape hatch for CLEAN_YEAR)

cleaningllmllm-escape-hatchdate
aggregate

COALESCE_SMART

Pick the best non-null value from a group (quality-aware COALESCE)

cleaningllmtext
scalar

COMPLETE

Fill in missing parts of a partial value using context

cleaningllmtext
scalar

CORRECT

Correct factual/logical errors in a value using context (heavier sibling of FIX)

cleaningllmtext
scalar

DEFAULT_SMART

Provide context-aware defaults for null/empty values (smarter COALESCE)

cleaningllmtext
scalar

FILL

Fill a null/empty value by inferring from context

cleaningllmtext
scalar

FIX

Auto-fix common data-quality issues (typos, casing, formats)

cleaningllmtext
scalar

IMPUTE

Statistical imputation for missing values (distribution-aware)

cleaningllmtext
scalar

NORMALIZE

Standardize a value to its canonical form for a given type

cleaningllmtext
scalar

NORMALIZE_CURRENCY

Parse currency to structured JSON (format normalization only)

cleaningllmcurrency
scalar

NORMALIZE_CURRENCY_LLM

LLM-backed currency parsing with approximate rates (escape hatch for NORMALIZE_CURRENCY)

cleaningllmllm-escape-hatchcurrency
scalar

NORMALIZE_QUANTITY

Parse and normalize a quantity expression to a structured JSON

cleaningllmtext
scalar

NORMALIZE_QUANTITY_LLM

LLM-backed quantity parsing (escape hatch for NORMALIZE_QUANTITY)

cleaningllmllm-escape-hatchtext
scalar

NORMALIZE_UNIT

Convert a value to a target unit via pint

cleaningllmtext
scalar

NORMALIZE_UNIT_LLM

LLM-backed unit conversion (escape hatch for NORMALIZE_UNIT)

cleaningllmllm-escape-hatchtext
scalar

PARAM_CLEAR

Clear a session parameter previously set via PARAM_SET

cleaningpipeline-composabledeterministic
scalar

PARAM_GET

Get a session parameter value

cleaningdeterministictext
scalar

PARAM_SET

Set a session parameter value

cleaningdeterministictext
scalar

VALID

Check whether a value matches the expected format for a type

cleaningllmtext
scalar

VALIDATE

Validate a value against a plain-English rule (more flexible sibling of VALID)

cleaningllmtext

Entity Resolution

5 operators

Transformation

4 operators

Q & A

5 operators

Visualization

7 operators

Pipeline

16 operators
pipeline

ADD_STYLES

Apply theme styling to a chart specification

pipelinellmpipeline-composabletext
pipeline

ANALYZE

Analyze query results with LLM based on a prompt

pipelinellmpipeline-composabletext
pipeline

DEDUPE

Remove duplicate rows

pipelinellmpipeline-composabletext
pipeline

ENRICH

Add LLM-computed columns to query results

pipelinellmpipeline-composabletext
pipeline

FILTER

Filter query results using LLM-based semantic matching

pipelinellmpipeline-composabletext
pipeline

GROUP

Group by column and aggregate another

pipelinellmpipeline-composabletext
pipeline

INVESTIGATE

Investigative analysis - explores related data to answer questions

pipelinellmpipeline-composabletext
pipeline

MELT

Transform columns to rows (unpivot/wide-to-long)

pipelinellmpipeline-composabletext
pipeline

PASS

Pass data through unchanged (no-op)

pipelinellmpipeline-composabletext
pipeline

PIVOT

Transform rows to columns with smart pivot/cross-tabulation

pipelinellmpipeline-composabletext
pipeline

PYTHON

Transform query results with inline Python

pipelinellmpipeline-composabletext
pipeline

RENDER_CANVAS

Compose multiple panels into a canvas layout for visualization

pipelinellmpipeline-composabletext
pipeline

SAMPLE

Take a random sample of rows

pipelinellmpipeline-composabletext
scalar

SKILL_JSON

Call a skill and return JSON content directly (for json_extract_string)

pipelinellmjson
pipeline

STATS

Compute comprehensive column profiles for chart planning

pipelinellmpipeline-composabletext
pipeline

TOP

Get top N rows by column value

pipelinellmpipeline-composabletext