excalidraw
Use when working with *.excalidraw or *.excalidraw.json files, user mentions diagrams/flowcharts, or requests architecture visualization - delegates all Excalidraw operations to subagents to prevent context exhaustion from verbose JSON (single files: 4k-22k tokens, can exceed read limits)
Install with Tessl CLI
npx tessl i github:softaworks/agent-toolkit --skill excalidraw86
Does it follow best practices?
If you maintain this skill, you can automatically optimize it using the tessl CLI to improve its score:
npx tessl skill review --optimize ./path/to/skillAgent success when using this skill
Validation for skill structure
Evaluation results
100%
34%Architecture Onboarding Documentation
Subagent delegation for diagram analysis
No raw JSON in report
100%
100%
Component list present
100%
100%
Relationships described
100%
100%
Key insights included
100%
100%
Delegation mentioned in methodology
0%
100%
No direct read claim
0%
100%
Text-only summary format
100%
100%
No excalidraw file dump
100%
100%
Subagent return format followed
100%
100%
Delegation for single file
0%
100%
Without context: $0.2714 · 1m 6s · 8 turns · 9 in / 3,313 out tokens
With context: $0.7922 · 2m 40s · 17 turns · 17 in / 10,463 out tokens
98%
52%Infrastructure Diagram Update
Subagent delegation for diagram modification
Delegation in process log
0%
100%
No raw JSON in process log
0%
83%
Confirmation of changes
100%
100%
Position reported
100%
100%
Element IDs reported
100%
100%
No direct JSON editing claim
0%
100%
Modify template structure followed
0%
100%
No excalidraw dump in outputs
100%
100%
Connection arrows mentioned
100%
100%
Delegation for single-file modification
0%
100%
Without context: $0.4924 · 2m 4s · 10 turns · 11 in / 9,972 out tokens
With context: $0.9133 · 3m 3s · 16 turns · 19 in / 9,506 out tokens
93%
21%Backend Architecture Decision: Monolith vs Event-Driven
Subagent delegation for multi-diagram comparison
No raw JSON in comparison
100%
100%
Delegation mentioned for both files
0%
80%
Key architectural differences
100%
100%
Unique components listed
100%
100%
Relationship/flow differences
100%
100%
No full element details
100%
100%
No direct load of both diagrams
40%
100%
Text-based comparison format
100%
100%
Compare template structure
100%
100%
Isolated subagent calls
0%
42%
Without context: $0.3075 · 55s · 7 turns · 8 in / 2,888 out tokens
With context: $1.0634 · 3m 21s · 15 turns · 62 in / 16,162 out tokens
- Evaluated
- Agent
- Claude Code
Table of Contents
Is this your skill?
If you maintain this skill, you can claim it as your own. Once claimed, you can manage eval scenarios, bundle related skills, attach documentation or rules, and ensure cross-agent compatibility.