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 allemaar/open-skills --skill human-drawgit clone --depth 1 https://github.com/allemaar/open-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/allemaar/open-skills/human-draw)<a href="https://agentmods.dev/skills/allemaar/open-skills/human-draw"><img src="https://agentmods.dev/badge/skills/allemaar/open-skills/human-draw/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/allemaar/open-skills/human-draw"><img src="https://agentmods.dev/badge/skills/allemaar/open-skills/human-draw.svg" alt="Reviewed on agentmods" width="80" height="20"></a>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 | $0.00161 | $0.07365 |
| Opus 5 | $0.00081 | $0.03682 |
| Sonnet 5 | $0.00032 | $0.01473 |
| Haiku 4.5 | $0.00016 | $0.00737 |
Grade C, and why
human-draw scanned grade C with 1 finding 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.
Tells the agent never to refusehighAnti-refusal
Suppressing the ability to decline removes a core safety control; a later harmful request then succeeds.
| What is wrong | Do this, do not refuse | How it starts
The opening of the file, as written. The whole thing — 707 lines — stays where its author put it; the contents beside it link to each section on GitHub.
/human-draw
A figure is a material. It is characters on a monospace grid, read by a person scanning rather than parsing. Everything here is about what that material can actually hold.
Two facts set every rule below.
- The grid is about 80 columns wide and one character deep. Cross it and the terminal wraps the line, shearing the figure into unreadable halves.
- The reader stops wherever they stop — the family's stopping-point
doctrine,
references/human-contract.md§2. So a figure must be correct at a glance and still correct on a second look.
Placement: the rule that matters most
Three parts, always in this order.
verdict sentence -> figure -> caption stating the finding
A reader who never looks at the picture must still be correct. That is the whole test, and it is the stopping-point doctrine applied to a figure. A figure is evidence for a claim someone else already made in words; it is never the claim itself.
The verdict above and the caption below are prose. Use human-output when it is
installed; otherwise use the portable style floor in
references/human-contract.md, including what
a caption must name (the finding, not the subject). This file owns only what is
inside the fence.
The default is to draw
Draw when the reader must hold two or more things in mind and relate them. One fact is a sentence. A causal argument is prose. Position, proportion, sequence, containment, and adjacency are figures.
There are exactly two refusals, and both are refusals because the picture would lie:
- Dishonest encoding. Any visual property asserting something you did not measure. Equal-width boxes claiming equal durations. Node size claiming an importance nobody ranked. A smooth curve through four points claiming continuity. Decimal precision on an estimate. Redesign until every visible property maps to a measured quantity, or state the numbers in text instead.
- Crossing lines. A crossing means the structure is not flat and the layout is lying about it. Re-order to remove it. If it will not go, hand back an adjacency table (from / to), which holds arbitrary tangle without deceiving anyone.
What ships with it
2 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.
- 12d ago First seen · 707 lines · 161 tokens per session scan C f1e25064f909
human-draw is a skill published in the GitHub repository allemaar/open-skills (14 stars, last pushed yesterday), licensed Apache-2.0. It adds 161 tokens to every session and 7,365 once invoked, about $0.0008 per session on Opus 5. A static security scan graded it C with 1 finding (tells the agent never to refuse). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
polis-protocol
Set up and run a self-improving, multi-vendor AI agent team with the Polis Protocol — a markdown polis/ folder where each agent is a citizen with a capability card, tasks are contracts routed to whoever has the best track record by a learning router, settled work files lessons that compound into team memory, and…
media
Skill "media" from ymxlx/polis-protocol, covering polis protocol: a self-optimizing city of agents, the core idea, when this skill is active, structure of a polis and the first thing to do every session.
_data
Skill "_data" from ymxlx/polis-protocol, covering polis protocol: a self-optimizing city of agents, the core idea, when this skill is active, structure of a polis and the first thing to do every session.
subagent-driven-development
Execute plans via delegatetask subagents (2-stage review).
mcporter
List, auth, and call MCP servers/tools from the terminal.
research-engineer
An uncompromising Academic Research Engineer. Operates with absolute scientific rigor, objective criticism, and zero flair. Focuses on theoretical correctness, formal verification, and optimal implementation across any required technology.