“Roman Empire around 100 AD”
Place the topic in world context with a focus map and who held power where.
← Back to projects
AILC History
Free Agent skill · MIT licensed
AILC History
AI Learning Companion.
An agent skill for interactive history tutoring grounded in the Wolfram Knowledgebase via Wolfram MCP — maps, timelines, conflict graphs, period art, and sticky stories.
Gallery
Examples from a session.
What it does
History that sticks — without inventing the map.
Open-ended questions about empires, wars, people, and periods become world context, dated borders, timelines, and carefully chosen artifacts — not dry lists, and not hallucinated map URLs.
- World context first — when, where at scale, who held power, who was at war.
- Maps & timelines from Wolfram — dated borders, battles, person lifespans.
- Primary sources & period art — coins, sources, and labeled AI reconstructions only when useful.
- Fast chat — high-level answer first; heavy Wolfram work in parallel background tasks.
- Honest about gaps — missing data, contested historiography, myth vs record.
Usage
Ask like a curious learner.
Trigger when exploring history — or use /ailc-history.
“Timeline of the Thirty Years’ War”
Build a dated conflict arc with actors and turning points from Wolfram entities.
“Map of the Ottoman Empire in 1520”
Compute borders for the year — never invent a map image URL.
Typical turn: immediate plain-language framing → parallel lookups (geopolitics, conflicts, map/timeline, era & sources) → one strong visual, a short story or quote, and a next hook.
Wolfram entity types
Compute over guess.
| Type | Use for |
|---|---|
HistoricalCountry |
Kingdoms, empires, dated borders |
MilitaryConflict |
Wars, battles, actor graphs |
HistoricalPeriod |
Eras, dynasties, cultural ages |
Person |
Notable people and lifespans |
HistoricalEvent |
Events, inventions, milestones |
GeographicRegion |
Continent-scale maps |
Names are resolved with free-form interpreters — never hand-written canonical entity IDs.
Requirements
What you need to run it.
Wolfram MCP handles entity resolution, maps, timelines, and conflict graphs. If Wolfram is unavailable, the skill falls back to careful prose and does not invent map image URLs.
made-to-stick is optional for SUCCESS storytelling technique; the skill still works without it.
Install
Clone into your agent’s skills path.
git clone https://github.com/warmersun/ailc-history-skill.git
# then link or copy into your skills directory
# (name as required by your host, e.g. ailc-history)
Ensure the host can load SKILL.md and has
Wolfram MCP configured (WolframContext,
WolframAlpha,
WolframLanguageEvaluator or equivalent).
Why we built it
Learning history should be a fun experience.
Maps you can trust, timelines that click, and stories that stick — not dry lists and invented URLs. AILC History is free and open source, from the people behind Warmer Sun.