Loading skill
Install any skill in seconds. Free to start, no credit card required.
Get Started Free →Generates Mermaid and ASCII diagrams of palace structure, knowledge topology, and synapse connectivity. Use when inspecting or presenting a palace visually.
.claude/skills/athola-palace-diagram/SKILL.md| Test case | Without → With | Effect | Δ tokens | Δ turns |
|---|---|---|---|---|
| case-01 | ✗→✓ | ▲ Improved | -1% | 0% |
| case-02 | ✗→✓ | ▲ Improved | -12% | 0% |
| case-05 | ✗→✓ | ▲ Improved | 283% | 0% |
| case-06 | ✗→✓ | ▲ Improved | -19% | 0% |
| case-13 | ✗→✓ | ▲ Improved | -35% | 0% |
Generate Mermaid and ASCII diagrams from the knowledge graph, showing palace structure, entity relationships, synapse connectivity, and tier assignments.
> Status: wired. Invoke via /palace diagram <palace-id> > [--type map|ascii|entity|heatmap], which calls > Skill(memory-palace:palace-diagram). The renderer lives in > memory_palace.palace_renderer.PalaceRenderer; see the Usage > section below for direct programmatic access.
| Type | Format | Description | |------|--------|-------------| | Palace map | Mermaid flowchart | Rooms as subgraphs, entities as nodes, synapses as edges | | Entity graph | Mermaid flowchart | Single entity's connections and triples | | Synapse heatmap | Mermaid flowchart | Edge styling by strength (thick=strong, dotted=weak) | | ASCII overview | Text | Box-drawing palace layout with entity counts |
PalaceRendererdisplay inline (for ASCII)
pythonfrom memory_palace.knowledge_graph import KnowledgeGraph from memory_palace.palace_renderer import PalaceRenderer graph = KnowledgeGraph("path/to/knowledge_graph.db") renderer = PalaceRenderer(graph) mermaid = renderer.palace_map("palace_id")
Then call mcp__claude_ai_Mermaid_Chart__validate_and_render_mermaid_diagram with the generated Mermaid string.
pythonascii_art = renderer.ascii_overview("palace_id") print(ascii_art)
pythonmermaid = renderer.entity_graph("entity_id")
pythonmermaid = renderer.synapse_heatmap("palace_id")
| Strength | Style | Meaning | |----------|-------|---------| | >= 0.7 | ==> (thick) | Strong connection | | >= 0.4 | --> (normal) | Medium connection | | < 0.4 | -.-> (dotted) | Weak connection |
Works with:
memory-palace-architect: visualize after palace creationknowledge-locator: display search results as graphgraph-analyzer: tier-informed node sizingsynapse heatmap, ASCII overview) is generated without a Python exception from PalaceRenderer
mcp__claude_ai_Mermaid_Chart__validate_and_render_mermaid_diagram and render successfully
entity counts per room
==> forstrength ≥ 0.7, --> for ≥ 0.4, -.-> for < 0.4
an error is reported with the palace ID that was searched
| Case | Status | Duration (ms) | Turns | Tokens | Tool calls | ||||||||
|---|---|---|---|---|---|---|---|---|---|---|---|---|---|
| Without | With | Δ | Without | With | Δ | Without | With | Δ | Without | With | Δ | ||
case-11 | pass→pass | 15,995 | 7,411 | -54% | 1 | 1 | 0% | 2,988 | 2,112 | -29% | 0 | 0 | — |
case-12 | pass→fail | 7,464 | 2,456 | -67% | 1 | 1 | 0% | 1,240 | 1,297 | +5% | 0 | 0 | — |
case-01 | fail→pass | 16,124 | 11,441 | -29% | 1 | 1 | 0% | 3,108 | 3,077 | -1% | 0 | 0 | — |
case-02 | fail→pass | 20,228 | 13,296 | -34% | 1 | 1 | 0% | 4,126 | 3,620 | -12% | 0 | 0 | — |
case-03 | fail→fail | 8,908 | 25,305 | +184% | 1 | 1 | 0% | 1,826 | 3,264 | +79% | 0 | 0 | — |
case-04 | fail→fail | 8,698 | 10,448 | +20% | 1 | 1 | 0% | 1,590 | 2,983 | +88% | 0 | 0 | — |
case-05 | fail→pass | 4,740 | 8,582 | +81% | 1 | 1 | 0% | 647 | 2,481 | +283% | 0 | 0 | — |
case-06 | fail→pass | 12,159 | 5,498 | -55% | 1 | 1 | 0% | 2,160 | 1,747 | -19% | 0 | 0 | — |
case-07 | pass→pass | 11,360 | 2,707 | -76% | 1 | 1 | 0% | 1,877 | 1,363 | -27% | 0 | 0 | — |
case-08 | pass→pass | 10,010 | 2,173 | -78% | 1 | 1 | 0% | 1,648 | 1,209 | -27% | 0 | 0 | — |
case-09 | pass→pass | 9,641 | 2,488 | -74% | 1 | 1 | 0% | 1,793 | 1,256 | -30% | 0 | 0 | — |
case-10 | pass→pass | 5,278 | 8,117 | +54% | 1 | 1 | 0% | 869 | 2,319 | +167% | 0 | 0 | — |
case-13 | fail→pass | 13,130 | 3,822 | -71% | 1 | 1 | 0% | 2,526 | 1,641 | -35% | 0 | 0 | — |
case-14 | fail→pass | 10,584 | 1,872 | -82% | 1 | 1 | 0% | 1,636 | 1,184 | -28% | 0 | 0 | — |
case-15 | fail→pass | 10,339 | 1,909 | -82% | 1 | 1 | 0% | 1,664 | 1,119 | -33% | 0 | 0 | — |
case-16 | fail→pass | 9,109 | 2,473 | -73% | 1 | 1 | 0% | 1,515 | 1,289 | -15% | 0 | 0 | — |
case-17 | fail→pass | 15,591 | 2,002 | -87% | 1 | 1 | 0% | 2,670 | 1,168 | -56% | 0 | 0 | — |
case-18 | fail→fail | 11,427 | 2,177 | -81% | 1 | 1 | 0% | 1,284 | 1,219 | -5% | 0 | 0 | — |
case-19 | fail→pass | 7,201 | 1,999 | -72% | 1 | 1 | 0% | 1,099 | 1,176 | +7% | 0 | 0 | — |
case-20 | fail→pass | 7,297 | 1,947 | -73% | 1 | 1 | 0% | 1,261 | 1,125 | -11% | 0 | 0 | — |
case-21 | fail→pass | 5,344 | 3,380 | -37% | 1 | 1 | 0% | 888 | 1,403 | +58% | 0 | 0 | — |
case-22 | pass→pass | 6,261 | 2,449 | -61% | 1 | 1 | 0% | 868 | 1,299 | +50% | 0 | 0 | — |
DecimalAI ran this skill against gemini-3.6-flash twice over the same eval suite — once with the skill loaded and once without — and compared the two runs case by case. 22 cases were attempted. The headline lift of +50 percentage points is the difference between those two pass rates over the 22 comparable cases. 1 case got worse with the skill loaded, and it is included in that figure.
Without the skill loaded, the model failed this case. With it loaded, the same prompt on the same model passed. This is one improved case from the latest verified run; every case, including any that regressed, is in the table above.
Other measured skills in the registry, with their headline benchmark lift.