# RinDig/Interpretable-Context-Methodology

Folder structure as agent architecture.  ICM replaces framework-level orchestration with filesystem structure.

Repository: https://github.com/RinDig/Interpretable-Context-Methodology
Canonical: https://ross.abutalabs.com/products/interpretable-context-methodology
Language: Python
License: MIT
License Family: permissive
Last push: 2026-07-25T16:17:00+00:00

## Health v2 (maintenance only)
Score: 57/100 (v2, computed 2026-09-02T17:46:02.011165+00:00)
- activity 94, release rhythm 35, longevity 13
- inputs: {"age_days": 192, "days_push": 39, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1087, forks 194 (observed 2026-08-28T04:03:32.412809+00:00)

## What it is
Interpretable Context Methodology (ICM) is a Python-based methodology that uses folder structure as agent architecture, replacing framework-level orchestration with filesystem organization. Numbered folders represent workflow stages, and markdown files carry prompts and context so a single AI agent can perform work that would otherwise require a multi-agent framework.

## Use cases
- orchestrate multi-step AI workflows without a multi-agent framework
- replace LangChain or CrewAI with a simpler filesystem-based pipeline
- build sequential agent workflows with human review at each stage
- edit prompts and context as plain markdown files instead of code
- run a single AI agent through staged tasks like research then writing
- set up a reusable workspace configuration for repeated content pipelines

## When to choose
- your workflow is sequential and benefits from human review between stages
- you want to modify pipeline steps by editing files rather than code
- you prefer plain-text, inspectable artifacts over framework abstractions
- you want one agent instead of maintaining a multi-agent setup

## When to avoid
- you need dynamic, non-sequential orchestration with branching or parallel agents
- you require built-in memory management, tool use, or error recovery from mature frameworks
- your pipeline stages need programmatic coordination or real-time communication

## Facets
- artifact type: framework
- maturity: active
- function: agent-framework, workflow-automation, prompt-engineering, llm-inference
- domain: large-language-models, developer-tools
- platform: python, cli, cross-platform
- tags: filesystem-orchestration, single-agent, markdown-prompts, unix-philosophy, multi-agent-alternative, human-in-the-loop, ai-agents, automation

## Member repositories
- RinDig/Interpretable-Context-Methodology (main) score 57

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:32.412809+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 2026-08-30T06:49:25.665953+00:00, confidence not recorded.
  - readme: https://github.com/RinDig/Interpretable-Context-Methodology (fetched 2026-08-28T04:03:32.412809+00:00, sha 46e1eebdaa4b)
- Data as of 2026-08-30T08:39:29.467469+00:00.