Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add naveedharri/benai-skills --skill excalidrawgit clone --depth 1 https://github.com/naveedharri/benai-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/naveedharri/benai-skills/excalidraw)<a href="https://agentmods.dev/skills/naveedharri/benai-skills/excalidraw"><img src="https://agentmods.dev/badge/skills/naveedharri/benai-skills/excalidraw/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/naveedharri/benai-skills/excalidraw"><img src="https://agentmods.dev/badge/skills/naveedharri/benai-skills/excalidraw.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 19 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
- medium Excessive Agency · line 25 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00099 | $0.04429 |
| Opus 5.5 | $0.00040 | $0.01772 |
| Sonnet 5.5 | $0.00020 | $0.00886 |
| Haiku 4.5 | $0.00010 | $0.00443 |
Grade A, and why
excalidraw scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 1mo ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 509 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Excalidraw Presentation Designer
Create compelling visual presentations through a collaborative, conversation-driven process. Every presentation is co-designed with the user — never assumed.
How This Skill Works
This is a guided conversation, not an assembly line. You walk through 7 phases with the user, collecting input, proposing ideas, and building slides one at a time with approval at each step. The goal is a presentation that feels like THEIRS, not a template.
Rules:
- Never build slides without understanding intent, audience, and motivation first
- Never batch-build all slides then reveal — build one, verify, iterate, then next
- Use
AskUserQuestionfor all structured decision points (2-4 options with descriptions) - Ask in small groups (2-3 questions max at a time) — this is a conversation, not a survey
- Don't proceed to the next phase without confirmation from the current one
Phase 0: Silent Config Load
Run this silently at the start of every session. Do NOT ask the user anything yet.
Step 0.1: Check for Saved Brand Config
Look for .excalidraw/brand.md in the working directory.
- If found: Read it silently. You'll use it in Phase 2.
- If not found: No action. You'll collect brand info in Phase 2.
Step 0.2: Check for Shared Context
Look for .claude/product-marketing-context.md in the working directory.
- If found: Read it silently. Extract any relevant company info, tone, audience details to inform later phases.
- If not found: No action.
Step 0.3: Acknowledge (briefly)
If either config was found, mention it in one line:
"I found your saved brand config — I'll use it unless you want to start fresh."
If neither was found, say nothing — proceed directly to Phase 1.
Phase 1: Understand Intent
Ask 2-3 structured questions to understand what the user needs. Use AskUserQuestion for each.
Question 1: What are you creating?
This determines the narrative arc suggestion in Phase 3.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 1mo ago Changed · +1 lines df49980fe858
- 2mo ago First seen · 508 lines · 99 tokens per session scan A 4f64af79b6f7
excalidraw is a skill published in the GitHub repository naveedharri/benai-skills (78 stars, last pushed 28d ago), licensed MIT. It adds 99 tokens to every session and 4,429 once invoked, about $0.0004 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-05.
excalidraw compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| excalidrawnaveedharri/benai-skills | Skill | 78 | 28d ago | MIT | A | 99 | Claude Code |
| doc-pptxmodu-ai/moai-cowork | Skill | 315 | 29d ago | Apache-2.0 | A | 266 | Claude Code |
| design-iteration-loopmodu-ai/moai-cowork | Skill | 315 | 29d ago | Apache-2.0 | A | 136 | Claude Code |
| design-landing-motionmodu-ai/moai-cowork | Skill | 315 | 29d ago | Apache-2.0 | A | 315 | Claude Code |
| design-logomodu-ai/moai-cowork | Skill | 315 | 29d ago | Apache-2.0 | A | 303 | Claude Code |
| design-system-librarymodu-ai/moai-cowork | Skill | 315 | 29d ago | Apache-2.0 | A | 234 | Claude Code |
| design-prompt-buildermodu-ai/moai-cowork | Skill | 315 | 29d ago | Apache-2.0 | A | 95 | Claude Code |