argument-visualization
Argument Visualization generates a color-coded mermaid diagram mapping claim relationships for a research topic by querying graph connections and writing structured wiki documentation. Use this skill when analyzing complex arguments to see how evidence supports or contradicts claims, identify logical structure, and surface the strongest and weakest positions in a debate.
git clone --depth 1 https://github.com/yogsoth-ai/de-anthropocentric-research-engine /tmp/argument-visualization && cp -r /tmp/argument-visualization/skills/argument-visualization ~/.claude/skills/argument-visualizationSKILL.md
# Argument Visualization
Generate a visual representation of the argument structure for a topic.
## Tool
`vault_query_graph` + CC file write
## Protocol
1. Query graph starting from the topic node, traversing supported_by, contradicts, and derived_from edges
2. Collect all claims, evidence, and their relationships
3. Format as a mermaid diagram in the wiki page:
- Green nodes: strong claims (strength ≥ 7)
- Yellow nodes: moderate claims (strength 4-6)
- Red nodes: weak claims (strength ≤ 3)
- Solid edges: supported_by
- Dashed edges: contradicts
4. Write/update `wiki/topics/<topic-slug>.md` with the argument map section
5. Include a summary: total claims, strongest/weakest, key contradictions
## HARD-GATE
<HARD-GATE>
Visualization must include at least 3 claims and their relationships. A single-node diagram is not useful.
</HARD-GATE>
## Yield
Returns: `{ topic: string, claims_shown: number, edges_shown: number, format: "mermaid" }`Experiment-specific - summarize the DARE executor's research design into a clean research_result report, forced to write back into the spec file produced by formated-specs.
Experiment-specific - replaces writing-specs, emits DARE's 4-layer call plan as a clean research_graph schema. Last step forces load formated-result.
loss-1 judge - read a sample's full dialogue and decide whether the user simulator semantically enacted its Policy Card. check-blind.
loss-2 judge - pairwise quality comparison across the n rungs within one topic; decide monotonicity and endpoint separation. check-blind, D1-D5 only.
Strategy: Inference to the best explanation in the face of anomalies
Remove components one by one, observe system changes to reveal hidden
Map system architecture to ablatable units for ablation studies
Design ablation studies to isolate component contributions in ML systems