---
type: Attested Computation
title: prompt
identifier: llm/prompt
description: 'The two files an LLM step reads, for llm/cohere''s prompt and system: prompt.md (the task, its subject, reference text, constraints) and system.md (the voice or house style, examples). Reference text and examples are length-capped and fenced as material, never instructions, so an earlier step''s output (headlines, a transcript, a page) cannot rewrite the task. For poems, poem/prompt knows the forms.'
pack: llm
tags:
- internet
runtime: fragletc
image: ofthemachine/python3@sha256:a015b4f9f0e7c1648f8fcd4164f6a296dd3c9189404f88f333a9764d08d13522
network: none
stdin: none
execution_class: hermetic
procedure_hash: sha256:3c7190b37c30ed1f113cdfde78f4bfe6a596e244d4ff10c07304c9d387642a0f
parameters:
- name: task
  type: string
  required: true
  description: What the model must produce, in one or two sentences
- name: subject
  type: string
  required: false
  description: What it is about, with the concrete names, numbers and events to keep
- name: context
  type: string
  required: false
  description: Reference text to work from, inline (often an earlier step's output); context_file for a file
- name: context_file
  type: file
  required: false
  description: Reference text to work from, as a file (e.g. news/headlines' headlines.md)
- name: context_label
  type: string
  required: false
  default: Reference material
  description: What the reference text is, named in prompt.md
- name: constraints
  type: string
  required: false
  description: Hard requirements such as length, form or things to avoid; newline- or semicolon-separated
- name: style
  type: string
  required: false
  description: The voice or house style for system.md, inline; style_file for a file
- name: style_file
  type: file
  required: false
  description: The voice or house style for system.md, as a file
- name: examples_file
  type: file
  required: false
  description: Examples of the wanted output (their voice is borrowed, never their content)
computation: https://tools.ofthemachine.com/llm/prompt.py
executor:
  resource: https://github.com/ofthemachine/fraglet
  receipt:
  - schema
  - fragletc
  - procedure
  - procedure_hash
  - image
  - image_digest
  - network
  - stdin_mode
  - params
  - inputs
  - argv
  - env
  - memo_key
  - started
  - finished
  - exit_code
  - stdout
  - outputs
  format: fraglet-receipt/2 (fragletc --receipt <path>, or FRAGLETC_RECEIPT_DIR)
attester:
  resource: https://tools.ofthemachine.com/receipt/attest.py
  method: 'deterministic, no re-execution: procedure_hash equals sha256 of computation; memo_key recomputes from the receipt''s own fields; claimed params and any artifact hash match the receipt'
generated:
  by: process:tools/catalog/compile.py
  at: '2026-11-13T18:27:40Z'
outputs:
- prompt.md
- system.md
---

# prompt

The two files an LLM step reads, for llm/cohere's prompt and system: prompt.md (the task, its subject, reference text, constraints) and system.md (the voice or house style, examples). Reference text and examples are length-capped and fenced as material, never instructions, so an earlier step's output (headlines, a transcript, a page) cannot rewrite the task. For poems, poem/prompt knows the forms.

Use as the step before llm/cohere when a request combines a task with text an earlier step produced or with a stored style guide, or whenever the prompt itself should be its own reproducible step.

# Schema

## Parameters
- `task` (string, required) — What the model must produce, in one or two sentences
- `subject` (string, optional) — What it is about, with the concrete names, numbers and events to keep
- `context` (string, optional) — Reference text to work from, inline (often an earlier step's output); context_file for a file
- `context_file` (file, optional) — Reference text to work from, as a file (e.g. news/headlines' headlines.md)
- `context_label` (string, optional, default `Reference material`) — What the reference text is, named in prompt.md
- `constraints` (string, optional) — Hard requirements such as length, form or things to avoid; newline- or semicolon-separated
- `style` (string, optional) — The voice or house style for system.md, inline; style_file for a file
- `style_file` (file, optional) — The voice or house style for system.md, as a file
- `examples_file` (file, optional) — Examples of the wanted output (their voice is borrowed, never their content)

## Outputs
- `prompt.md` (declared file under /output)
- `system.md` (declared file under /output)

# Execution

The computation is [`llm/prompt.py`](https://tools.ofthemachine.com/llm/prompt.py) (frontmatter `computation`; in `tools.tar.gz` the same file sits at `llm/prompt.py` beside this bundle), run by `fragletc` in `ofthemachine/python3@sha256:a015b4f9f0e7c1648f8fcd4164f6a296dd3c9189404f88f333a9764d08d13522` with network `none`. Its identity is `procedure_hash` = sha256 of that file, shebang included — the same value every receipt records. Parameters are passed as `-p name=value`; `--receipt run.json` (or `FRAGLETC_RECEIPT_DIR`) writes the receipt, a `fraglet-receipt/2` JSON document with the fields listed under `executor.receipt`.

# Attestation

A run's receipt is attested with [`receipt/attest.py`](https://tools.ofthemachine.com/receipt/attest.py): the computation that ran is this script (`procedure_hash`), the receipt is internally consistent (`memo_key` recomputes from its own fields), the claimed parameters are the ones that ran, and an artifact claimed as an output hashes to what the receipt recorded. No re-execution: attestation is provenance fidelity, not reproduction.

Hermetic: no network, so `memo_key` also fully determines the outputs — a re-run with the same inputs must reproduce them, which makes results from this tool safely memoizable.

Skill: [SKILL.md](https://tools.ofthemachine.com/llm/SKILL.md)
