# rasbt/mini-coding-agent

Minimal and readable coding agent harness implementation in Python to explain the core components of coding agents.

Repository: https://github.com/rasbt/mini-coding-agent
Canonical: https://ross.abutalabs.com/products/mini-coding-agent
Homepage: https://magazine.sebastianraschka.com/p/components-of-a-coding-agent
Language: Python
License: Apache-2.0
License Family: permissive
Topics: agents, ai, large-language-models, llms, python
Last push: 2026-04-07T22:07:26+00:00

## Health v2 (maintenance only)
Score: 49/100 (v2, computed 2026-09-03T02:39:23.370411+00:00)
- activity 76, release rhythm 35, longevity 11
- inputs: {"age_days": 154, "days_push": 148, "days_rel": null, "gap_med": null, "n_releases_24m": 0}
- flags: no_releases, young
- formula: round(0.45*activity + 0.35*rhythm + 0.20*longevity); archived -> min(score, 10)

## Adoption (not part of the score)
Stars 1126, forks 203 (observed 2026-08-28T04:03:41.310527+00:00)

## What it is
A minimal, readable Python implementation of a coding agent harness that demonstrates the six core components of coding agents (repo context, prompt stability, structured tools, context management, memory, and delegation). It accompanies Sebastian Raschka's tutorial 'Components of a Coding Agent' and runs locally via Ollama with no dependencies beyond the standard library.

## Use cases
- learn how coding agents like Claude Code work internally
- understand agent harness components like tool use and memory
- build my own coding agent from scratch
- study a minimal agent loop implementation in Python
- run a local coding agent with Ollama
- teach LLM agent architecture concepts
- see how prompt caching and context management work in agents

## When to choose
- you want to learn the internals of coding agents from a small readable codebase
- you prefer a dependency-free Python implementation running on local Ollama models
- you need teaching material or a reference for agent harness design
- you want a starting point to experiment with agent components like approvals, transcripts, and subagents

## When to avoid
- you need a production-grade coding assistant comparable to Claude Code or Codex CLI
- you require cloud LLM provider support beyond Ollama
- you want a feature-rich agent with extensive tooling and integrations
- you need multi-user or team deployment

## Facets
- artifact type: learning-resource
- maturity: active
- function: agent-framework, cli, llm-inference, developer-tools
- domain: large-language-models, developer-tools, tutorials
- platform: python, cli, windows
- tags: coding-agent, agent-harness, ollama, educational, minimal-implementation, tool-use, prompt-caching, subagents, ai-agents, command-line, linux, macos

## Member repositories
- rasbt/mini-coding-agent (main) score 49

## Provenance
- Observed fields: from GitHub, fetched 2026-08-28T04:03:41.310527+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 2026-08-30T06:39:13.386608+00:00, confidence not recorded.
  - readme: https://github.com/rasbt/mini-coding-agent (fetched 2026-08-28T04:03:41.310527+00:00, sha d9c6174ab28e)
  - homepage: https://magazine.sebastianraschka.com/p/components-of-a-coding-agent (fetched 2026-08-29T12:43:49.680551+00:00, sha 454d3957879b)
- Data as of 2026-08-30T08:39:29.467469+00:00.