# openprose/prose

A new kind of language for a new kind of computer

Repository: https://github.com/openprose/prose
Canonical: https://ross.abutalabs.com/products/openprose-prose
Homepage: https://prose.md
Language: TypeScript
License: MIT
License Family: permissive
Last push: 2026-08-26T01:21:08+00:00

## Health v2 (maintenance only)
Score: 82/100 (v2, computed 2026-09-03T02:20:16.233290+00:00)
- activity 99, release rhythm 97, longevity 17
- inputs: {"age_days": 242, "days_push": 8, "days_rel": 21, "gap_med": 1, "n_releases_24m": 16}
- flags: none
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1688, forks 136 (observed 2026-08-28T04:05:22.129572+00:00)

## What it is
OpenProse is a declarative programming language for standing AI agent work, authored as Markdown contracts (*.prose.md) that declare desired world states rather than instruction sequences. It ships as a skill installed into any Prose-Complete agent harness (Claude Code, Codex CLI, OpenCode), where the agent session itself acts as the runtime that reconciles reality with the declared contracts.

## Use cases
- declare standing AI jobs as markdown contracts instead of scripting agent steps
- keep a live risk report current across changing data with an agent
- define typed AI functions with parameters, returns, and tool requirements
- run declarative agent workflows inside Claude Code or Codex CLI
- reconcile an ideal world state automatically when inputs change
- add imperative control flow to agent tasks only where order matters
- install an agent language skill into an existing coding agent

## When to choose
- you want standing, continuously-maintained AI outcomes rather than one-shot prompt scripts
- you prefer declarative, reconciled agent behavior in the style of Terraform or Kubernetes
- you already use a Prose-Complete harness like Claude Code or Codex CLI and want a shared contract language

## When to avoid
- you need a standalone binary or traditional compiler/runtime independent of an LLM agent host
- you want deterministic, non-model execution with guaranteed latency or cost bounds
- you need a mature, production-hardened language rather than an early experimental design

## Facets
- artifact type: cli-tool
- maturity: experimental
- function: agent-framework, prompt-engineering, workflow-automation, configuration-management, parser, sdk
- domain: artificial-intelligence, developer-tools, programming-languages, large-language-models
- platform: cli, cross-platform
- tags: declarative-language, markdown-contracts, ai-agent-harness, prose-complete, skills, reconciliation, standing-jobs, prosescript, llm-runtime, programming-language, ai-agents, automation, nodejs

## Member repositories
- openprose/prose (main) score 82

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:05:22.129572+00:00.
- Health v2: computed from the inputs above; adoption is never an input.
- Inferred fields (summary, facets, guidance): AI-extracted, prompt v1, taxonomy v1, on 2026-08-30T08:32:31.150886+00:00, confidence not recorded.
  - readme: https://github.com/openprose/prose (fetched 2026-08-28T04:05:22.129572+00:00, sha 6eea50446278)
  - homepage: https://prose.md (fetched 2026-08-29T11:14:01.046027+00:00, sha 2d8983ee4b9e)
  - site_page: https://docs.prose.md/ (fetched 2026-08-29T11:14:01.055346+00:00, sha f96d78b794c7)
  - site_page: https://docs.prose.md/start/what-is-openprose (fetched 2026-08-29T11:14:01.057240+00:00, sha b1ce4546f5cd)
- Data as of 2026-08-30T08:39:29.467469+00:00.