fp-pipe-ref – Data & Science skill for AI agents

pipe & flow Quick Reference

pipe - Transform a Value

import { pipe } from 'fp-ts/function'

// pipe(startValue, fn1, fn2, fn3)
// = fn3(fn2(fn1(startValue)))

const result = pipe(
  '  hello world  ',
  s => s.trim(),
  s => s.toUpperCase(),
  s => s.split(' ')
)
// ['HELLO', 'WORLD']

flow - Create Reusable Pipeline

import { flow } from 'fp-ts/function'

// flow(fn1, fn2, fn3) returns a new function
const process = flow(
  (s: string) => s.trim(),
  s => s.toUpperCase(),
  s => s.split(' ')
)

process('  hello world  ') // ['HELLO', 'WORLD']
process('  foo bar  ')     // ['FOO', 'BAR']

When to Use

| Use | When |

|-----|------|

| pipe | Transform a specific value now |

| flow | Create reusable transformation |

With fp-ts Types

import * as O from 'fp-ts/Option'
import * as A from 'fp-ts/Array'

// Option chain
pipe(
  O.fromNullable(user),
  O.flatMap(u => O.fromNullable(u.email)),
  O.getOrElse(() => 'no email')
)

// Array chain
pipe(
  users,
  A.filter(u => u.active),
  A.map(u => u.name)
)

Common Pattern

// Data last enables partial application
const getActiveNames = flow(
  A.filter((u: User) => u.active),
  A.map(u => u.name)
)

// Reuse anywhere
getActiveNames(users1)
getActiveNames(users2)

Limitations

• Use this skill only when the task clearly matches the scope described above.
• Do not treat the output as a substitute for environment-specific validation, testing, or expert review.
• Stop and ask for clarification if required inputs, permissions, safety boundaries, or success criteria are missing.

Reference scope

These are fp-ts 2.x fragments. Supply application variables, types, validators and imports, and check the installed version. map does not catch thrown exceptions. Task has no typed error channel; an improperly wrapped promise can still reject. Use tryCatch at fallible boundaries.

Official API: [fp-ts modules](https://gcanti.github.io/fp-ts/modules/).