Task agent prompt:
"Generate ISO 11179-compliant glossary definitions for the following [N] terms.
Write ALL entries as markdown (#### headers with definitions, examples, and
discussion) to the file /tmp/glossary-raw.md using the Write tool.
Each entry uses #### for the term header. Do not return the content in your
response — just confirm the file was written and report the term count.

[Paste the full term list here]

[Paste course description context here for audience level]"

undefined

undefined

#!/usr/bin/env python3
"""Merge glossary parts into a single sorted glossary."""
import glob, os, re

entries = {}

Run this script with `python3 /tmp/assemble_glossary.py` via the Bash tool.
Total cost: ~500 tokens for the script + ~200 tokens for output = **~700 tokens**
(versus 200,000+ tokens if done manually through Edit calls).

**NEVER DO ANY OF THE FOLLOWING:**

- Write glossary entries directly through the Write or Edit tool
- Copy-paste subagent output into Edit tool old_string/new_string parameters
- Manually sort terms by emitting them in alphabetical order
- Append sections to the glossary file one at a time through Edit calls

**Formatting rules for the assembled file:**

- Do not put any `---` strings in the glossary. They are not needed.
- Sort all terms alphabetically (case-insensitive) — the script handles this
- Use level-4 headers (####) for term names
- Place definition in body text (no special formatting)
- Use "**Example:**" for examples (bold, with colon)
- Maintain consistent spacing between entries (one blank line between entries)

{
  "terms": [
    {
      "term": "Learning Graph",
      "related_terms": ["Concept Dependency", "Directed Acyclic Graph"],
      "contrasts_with": ["Linear Curriculum"],
      "category": "Educational Technology"
    }
  ]
}