AI Summary Generation

AI Summary Generation Workflow

Generates concise, standardised summaries from Markdown files using a local Python script connected to the OpenAI API.

---

Prerequisites

• Python 3.9+ installed

• OPENAI_API_KEY set as an environment variable

• Dependencies installed (pip install -r requirements.txt)

• Source Markdown file located under docs/

---

Usage

Run from the repository root:

python scripts/chatgpt_summary.py docs/system/architecture.md

Optional parameters:

--model gpt-4o-mini
--max-tokens 800
--output-dir docs/system/summaries/

---

Process Summary

Generates a summarized version of the document while preserving metadata for provenance and version tracking.

---

Inputs

• Markdown file in docs/

• YAML metadata fields (title, tags, status, created, updated, version)

• Script path: scripts/chatgpt_summary.p

---

Outputs

• <filename>.summary.md created in the same directory as the source file

• Original YAML frontmatter is preserved with an additional generated: true field

• Included in auto-index.md on next regeneration

---

Example

Input:
docs/system/architecture.md

Output:
docs/system/architecture.summary.md

---

Notes

• Ensure the OpenAI API key is configured in your environment before execution.

• The generated file is committed manually, to maintain transparency in version control.

---