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 BV-Venky/excalidraw-architect-mcp --skill excalidraw-architectgit clone --depth 1 https://github.com/BV-Venky/excalidraw-architect-mcpWrote 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/bv-venky/excalidraw-architect-mcp/excalidraw-architect)<a href="https://agentmods.dev/skills/bv-venky/excalidraw-architect-mcp/excalidraw-architect"><img src="https://agentmods.dev/badge/skills/bv-venky/excalidraw-architect-mcp/excalidraw-architect/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/bv-venky/excalidraw-architect-mcp/excalidraw-architect"><img src="https://agentmods.dev/badge/skills/bv-venky/excalidraw-architect-mcp/excalidraw-architect.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 2 findings, up to high
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 →
- high Prompt Injection · line 48 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
- high Prompt Injection · line 48 Hidden instructions were detected in comments or invisible text. These could contain malicious directives. Manual review is recommended.Fix: Audit all comments and invisible characters. Remove any instructions that direct the agent to perform unauthorized actions. Use plain, reviewable content.
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 | $0.00101 | $0.01880 |
| Opus 5 | $0.00051 | $0.00940 |
| Sonnet 5 | $0.00020 | $0.00376 |
| Haiku 4.5 | $0.00010 | $0.00188 |
Grade A, and why
excalidraw-architect 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 12d 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 — 140 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Excalidraw Architect
The MCP server renders. This skill decides what to render and how much.
Picking the wrong diagram type is the most common way a diagram fails. A cross-functional process drawn as a flowchart loses the handoffs that were the entire point. A ranking drawn as a bar chart loses the ranking.
1. Before you draw
Ask: would the reader learn more from this than from a well-written paragraph or a three-column table? If no, write the paragraph.
Don't draw for:
- A list of things → bullets or a table.
- A simple before/after → a two-column table.
- One shape → just write the sentence.
2. The two rules that carry most of the quality
Density target: ~4/10. Enough to be complete, not so dense it needs its own legend. Above nine nodes it is usually two diagrams — draw an overview and a detail. Two nodes that always travel together are one node.
Accent is reserved. Every spec accepts "focal": true on its elements.
Mark one or two — the thing you want looked at first. Marking five spends
the signal and leaves the reader with no entry point. Most elements should be
unmarked; that is what makes the marked one work.
A third, quieter rule: every connection should carry information. If the relationship is already obvious from the layout, delete the line.
3. Picking a type
| If you're showing… | Use | Not when |
|---|---|---|
| Components and the connections between them | architecture |
(graph type — pass nodes + connections) |
| Decision logic with branches | flowchart |
(graph type — pass nodes + connections) |
| Parent -> children: taxonomies, file hierarchies, decision breakdowns. | tree |
Use org_chart for people or teams; use nested when containment is the point. |
| Human, team, or agent ownership: reporting lines, escalation, routing. | org_chart |
Not for system components -- that is architecture. |
| States, transitions, and guards. Order lifecycles, retry machines, approval flows. | state |
Not for time-ordered messages between actors -- that is sequence. |
| Hierarchy by containment or scope: bounded contexts, module trees, org scopes. | nested |
Beyond three levels of nesting, split into an overview plus a detail diagram. |
| Stacked abstraction levels where each layer sits on the one below. | layers |
If the layers exchange traffic in both directions, use architecture instead. |
| Multi-tier data storage with quality levels: bronze/silver/gold, raw/curated/serving. | medallion |
Use data_flow when the emphasis is who does what rather than where data rests. |
| Entities, their fields, and relationships. Data models and schema documentation. | er |
Above ~8 entities, split by bounded context. |
| End-to-end stack drawn inside one named boundary (a cluster, an account, a VPC). | high_level |
Use architecture when the boundary is not the point. |
| Legacy landscape grouped by phase or department; the 'before' state. | it_state |
Use architecture for a target-state design. |
| Cross-functional process where step ownership matters. Handoffs are the story. | swimlane |
If every step has the same owner, use flowchart. |
| Multi-actor sequential workflow drawn as vertical actor columns with data handoffs. | process |
Use swimlane for horizontal lanes and a wider flow. |
| A pipeline where each stage is tagged with the role that owns it. | data_flow |
Use medallion when storage tiers, not roles, are the subject. |
| Integration topology of a platform: sources -> core -> consumers. | dp_integration |
Use architecture when the core needs to be decomposed in detail. |
| Time-ordered messages between actors: API traces, protocols, incident replays. | sequence |
Never draw a message arrow that points backwards in time; use state for lifecycles. |
| Events positioned in time along a single axis. Roadmaps, incident chronologies. | timeline |
Use gantt when items have duration rather than a single instant. |
| Two-axis positioning and prioritization. Add quadrant_labels for a consultant 2x2. | quadrant |
Use scatter when the axes carry real units rather than relative position. |
| Ranked hierarchy (mode='pyramid') or conversion drop-off (mode='funnel'). | pyramid |
Use bar when the values matter more than the ranking. |
| Overlap between two or three sets, where the intersection is the insight. | venn |
Four or more sets is unreadable -- use a table. |
| A reinforcing cycle; add a hub for shared state the stations write back to. | loop |
Use process when the flow has a real beginning and end. |
| Tasks and phases with duration on a timeline. Shows overlap and parallel tracks. | gantt |
Above ~12 tasks, collapse to a phase-level view. |
| Quantitative comparison across categories. Multiple series render as grouped bars. | bar |
Use line when the x axis is continuous time. |
| Continuous trends over an ordered axis. Capped at 15 points per series. | line |
Use bar for a handful of discrete categories. |
| Distribution and correlation between two continuous variables. | scatter |
Use quadrant when position is relative judgement rather than measurement. |
What ships with it
1 file 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.
- 12d ago First seen · 140 lines · 101 tokens per session scan A 4e5ac3b991f2
excalidraw-architect is a skill published in the GitHub repository BV-Venky/excalidraw-architect-mcp (150 stars, last pushed 17d ago), licensed MIT. It adds 101 tokens to every session and 1,880 once invoked, about $0.0005 per session on Opus 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 2026-08-30.
Other skills, from other repositories
pn-image-creator
Questionnaire-driven image creation for high-quality PNG or SVG. Always grounds prompts in pn-cinematography-lighting and pn-image-prompt-engineering (camera, lighting, visual style). Gates on user confirmation before generation. Use with Cursor image generation or SVG per pn-svg-creator.
design
Comprehensive design skill: brand identity, design tokens, UI styling, logo generation (55 styles, Gemini AI), corporate identity program (50 deliverables, CIP mockups), HTML presentations (Chart.js), banner design (22 styles, social/ads/web/print), icon design (15 styles, SVG, Gemini 3.1 Pro), social photos…
design
Comprehensive design skill: brand identity, design tokens, UI styling, logo generation (55 styles, Gemini AI), corporate identity program (50 deliverables, CIP mockups), HTML presentations (Chart.js), banner design (22 styles, social/ads/web/print), icon design (15 styles, SVG, Gemini 3.1 Pro), social photos…
canvas-design
Create beautiful visual art in .png and .pdf documents using design philosophy. You should use this skill when the user asks to create a poster, piece of art, design, or other static piece. Create original visual designs, never copying existing artists' work to avoid copyright violations.
brandkit
Premium brand-kit image generation skill for creating high-end brand-guidelines boards, logo systems, identity decks, and visual-world presentations. Trained for minimalist, cinematic, editorial, dark-tech, luxury, cultural, security, gaming, developer-tool, and consumer-app brand systems. Optimized for intentional…
slides
Create strategic HTML presentations with Chart.js, design tokens, responsive layouts, copywriting formulas, and contextual slide strategies.