excalidraw
Generate an Excalidraw whiteboard in Sean's hand-drawn video style (Excalifont, roughness 1, green signature, socials + watermark, source labels).
Skill by gxiaopang
Chalk Labs
Need help deploying excalidraw?
Chalk Labs sets up agents like this inside real teams: connected to your tools, tested on your data, with a human in the loop where it matters.
Install
Works with Claude Code, Claude apps, Codex, Cursor, Gemini CLI, GitHub Copilot, VS Code, OpenCode, Goose, Windsurf. Read the repository before you give any tool access to your data.
npx agentmarket install skill/gxiaopang-excalidraw
Asks which of your tools to add it to (Claude Code, Codex, Cursor, VS Code, Windsurf, Gemini CLI, Claude Desktop) and shows the change first. See https://agentmart.dev/cli.
Get the weekly diff
The weekly diff: one email a week with new agent tools, notable updates and grade changes, plus security alerts for tools you follow. Double opt-in, one-click unsubscribe.
About
Use cases: data
Reviews
Write a reviewNo reviews yet. Installed it? Reviews come from verified installers only. How reviews work
Permissions
What this skill can do and what it asks for, from its SKILL.md. Checked Oct 25, 2026.
- Tools
- No tools found
- Secrets
- None asked for
- Install scripts
- None found
Network hosts it talks to 1
codeload.github.com· found in install command
Files it touches
~/.claude/skills/excalidraw
What's inside
- Skill name
excalidraw- Description
- Generate an Excalidraw whiteboard in Sean's hand-drawn video style (Excalifont, roughness 1, green signature, socials + watermark, source labels). Use whenever the user wants a whiteboard, diagram, teaching board, or "chart" for a video or the docs/whiteboards gallery or a lab/ topic — anything Sean will film with.
- License
- MIT
- Instructions
- 98 lines in SKILL.md · view on GitHub
Files (1)
- SKILL.md
SKILL.md (first lines)
# Excalidraw whiteboards, Sean's way Boards are generated from code so they match the masters in `~/Developer/Excalidraw` exactly — Sean films with them directly. Never hand-write `.excalidraw` JSON; use the style engine, which locks the values that a from-scratch board always gets wrong. ## The five things a naive board gets wrong (all fixed by the engine) 1. `fontFamily 5` (Excalifont) — NOT `1` (legacy Virgil). Wrong font = wrong board. 2. `roughness 1` (controlled) — NOT `2` (scratchy). 3. Stroke-width hierarchy `1` detail / `2` container / `4` emphasis — not flat `2`. 4. Type scale = Excalidraw presets S=16 / M=20 / L=28. Majority is S/M; L (28) reserved for big titles only. Don't inflate past the presets. 5. Signature green `#b2f2bb`, frames, socials block, watermark, source labels. ## How to build one ```python from scripts.whiteboard import style as S # run from the repo root e = [] e.append(S.text(60, 40, "Title", size=S.FS_TITLE)) e.append(S.underline(64, 120, 800, color=S.PAL["orange"][1])) e += S.labeled_box(80, 200, 300, 140, "A box\nwith a note", color="green") e += S.pill_header(80, 400, 600, "SECTION", color="blue") e += S.ellipse(500, 200, 180, 120, "agent", color="pink") e += S.diamond(700, 200, 140, 100, "gate?", color="green") e += S.labeled_arrow(380, 270, 500, 260, "tool calls") e += S.socials_block(1980, 44) e.append(S.watermark(80, 1030)) e.append(S.source_label(80, 900, "per <vendor>, <date>")) # standing rule e.append(S.red_note(80, 960, "honest red-ink opinion")) # standing rule S.validate(e) # catches broken bound-text ids doc = S.document(e) # wraps with appState + white bg ``` Where the board goes depends on who it is for: - **A board that explains Waku's own code:** write it to `docs/whiteboards/<name>.excalidraw`. - **A board for a video** (anything a `lab/` topic films): the `.excalidraw` source and the script that draws it go to `~/Developer/Excalidraw/waku-lab/<topic>/`, **never into the repo**. Commit only a PNG screenshot, to `lab/<topic>/screenshots/<name>.png`. `evals/deterministic/test_rulebook.py` fails if a `.excalidraw` file appears in a lab topic. See `lab/kimi-k3/build_k3_tutorial.py` for a full two-board example. ## Palette (name → fill/stroke), meaning follows Sean's color system - `green` loop / hero / final reply · `red` harness boundary / cost / honest-ink - `blue` LLM-ops / observability · `orange` a loop step · `pink` LLM/agent nodes - `grey` neutral state/config · `yellow` callout · `plain` bare container ## Always, before delivering 1. `S.validate(elements)` — every bound text must map to a container. 2. Eyeball via SVG proxy → PNG (QuickLook crops wide canvases; use headless Chrome `--screenshot --window-size=W,H` for the full board). Check for overlaps and off-canvas elements. 3. Shareable boards get the socials block + `@ShenSeanChen` watermark. 4. Every technical/vendor claim gets a dated `source_label`. ## Two ratios, measured from the masters — check yours before delivering Generated boards fail in the same two ways every time. Both are countable, so count them instead of eyeballing: ```python # against ~/Developer/Excalidraw/<newest>.excalidraw black strokes / all shapes -> masters: 96% (354 of 368) arrows / all shapes -> masters: 0.88 (241 arrows, 165 ellipses) ``` 1. **COLOUR POLLUTION.** The masters are ~96% BLACK strokes with transparent or white fills. Colour lands maybe fifteen times on a whole board: green `#b2f2bb` for the hero, red `#ffc9c9` for the warning, and a handful of coloured strokes on section boundaries. A first-pass generated board typically comes out at 3% black — a coloured stroke on every shape. Ink by default; a fill has to earn its place. Note `labeled_box(fill="chip")` is documented as user-prompt/reply chips ONLY, and using it for general boxes is the usual cause. 2. **NO FLOW.** The masters average nearly one arrow per shape, with ellipses for agents and diamonds for decisions. A board of boxes in a grid with two arrows is an infographic, not a whiteboard — it reads as dead. If the subject has motion, DRAW the motion: fan-outs, back-edges, branches from a diamond. A back-edge that visibly returns is worth more than a paragraph. Fix both before showing it, because they are the first things Sean notices. ## Layout lessons (from comparing generated vs. hand-drawn) - Poster-tight beats sprawl for a single teaching idea — fit it in one landscape frame (~2500×1100) so it reads without panning. Reserve the 8000×20000 sprawl for full system maps. - A diagonal "ladder" reads as progress; a red barrier + a green break-through reads as problem→solution. Reuse those two shapes.
From gxiaopang/ContractGuard-Agent, MIT.
More from gxiaopang/ContractGuard-Agent
- 0.561score
ip-ownership
gxiaopang
IP ownership assignment review procedures examine intellectual property transfers, background rights, retained licenses.
SkillAFreshness: AClaude Code +9 - 0.558score
cap-on-liability
gxiaopang
Cap on liability procedures examine aggregate ceilings, monetary limits, fee multiples, sublimits.
SkillAFreshness: AClaude Code +9 - 0.558score
change-of-control
gxiaopang
Change of control procedures examine ownership thresholds, mergers, acquisitions, consent triggers.
SkillAFreshness: AClaude Code +9 - 0.558score
confidentiality
gxiaopang
Confidentiality procedures examine obligations, confidential information, nondisclosure duties, disclosure exceptions.
SkillAFreshness: AClaude Code +9 - 0.558score
exclusivity
gxiaopang
Exclusivity review procedures examine exclusive rights, sole suppliers, distribution arrangements.
SkillAFreshness: AClaude Code +9 - 0.558score
indemnification
gxiaopang
Indemnification review procedures examine indemnity coverage, defense duties, covered claims, beneficiaries.
SkillAFreshness: AClaude Code +9 - 0.558score
non-compete
gxiaopang
Non-compete procedures examine restrictions, competing businesses, competitive activities, territorial scope.
SkillAFreshness: AClaude Code +9 - 0.558score
termination-for-convenience
gxiaopang
Termination for convenience procedures examine no-cause cancellation, notice periods, exit fees.
SkillAFreshness: AClaude Code +9 - 0.558score
uncapped-liability
gxiaopang
Uncapped liability procedures examine unlimited exposure, limitation carve-outs, damage exclusions.
SkillAFreshness: AClaude Code +9 - 0.558score
governing-law
gxiaopang
Governing law procedures examine choice-of-law provisions, jurisdiction selections, conflicting laws.
SkillAFreshness: AClaude Code +9 - 0.524score
review-pr
gxiaopang
Walk Sean through an incoming waku-agent PR or issue and present it his way — four fixed sections: what this is, why it matters, how HE can test it with you…
SkillBFreshness: BClaude Code +9 - 0.304score
research-report
gxiaopang
No description from the publisher yet.
SkillAFreshness: AClaude Code +9
Signals
- Stars
- 0
- Forks
- 0
- Open issues
- 0
- Last commit (this folder)
- Sep 17, 2026 (1 month ago)
From GitHub (fetched Oct 24, 2026).
Details
- License
- MIT
- Repository
- github.com/gxiaopang/ContractGuard-Agent
- First seen
- Oct 24, 2026
- Last checked
- Oct 24, 2026
- Source
- GitHubgithub-graphql
More from gxiaopang
- 0.561score
ip-ownership
gxiaopang
IP ownership assignment review procedures examine intellectual property transfers, background rights, retained licenses.
SkillAFreshness: AClaude Code +9 - 0.558score
cap-on-liability
gxiaopang
Cap on liability procedures examine aggregate ceilings, monetary limits, fee multiples, sublimits.
SkillAFreshness: AClaude Code +9 - 0.558score
change-of-control
gxiaopang
Change of control procedures examine ownership thresholds, mergers, acquisitions, consent triggers.
SkillAFreshness: AClaude Code +9 - 0.558score
confidentiality
gxiaopang
Confidentiality procedures examine obligations, confidential information, nondisclosure duties, disclosure exceptions.
SkillAFreshness: AClaude Code +9 - 0.558score
exclusivity
gxiaopang
Exclusivity review procedures examine exclusive rights, sole suppliers, distribution arrangements.
SkillAFreshness: AClaude Code +9 - 0.558score
indemnification
gxiaopang
Indemnification review procedures examine indemnity coverage, defense duties, covered claims, beneficiaries.
SkillAFreshness: AClaude Code +9
Done for you
Rather not set it up yourself?
Chalk Labs scopes, builds and runs agents for your team: the task audit, the eval harness, the integrations and the weekly ops report. Fixed scope, priced before we start.
See how we build AI agents →Read the case studies
Chalk Labs is a two-founder AI and growth studio. We build AI agents, CRMs and MVPs for Web3 and AI companies, and have shipped 50+ product and token launches.
- launches shipped
- 50+
- launches shipped
- media placements
- 500+
- media placements
- founders on every account
- 2
- founders on every account
Press we've placed: Forbes · CoinDesk · Decrypt · Cointelegraph · The Block
Rahil Jain, Co-founder, Chalk Labs. Ex-Goldman Sachs engineer. You talk to the person who builds it.
Get excalidraw running in your business
Tell us what you want to automate. A founder replies within one working day with a plan, not a pitch deck.
Cite this page
excalidraw skill. Agent Mart. https://agentmart.dev/skills/gxiaopang-excalidraw (last updated 2026-10-24)