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 agentmods add skills/br3eze-code/br3eze-code/agentos-draftsmannpx skills add br3eze-code/br3eze-code --skill agentos-draftsmangit clone --depth 1 https://github.com/br3eze-code/br3eze-codeWhat 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 | $0.00045 | $0.00387 |
| Opus 5 | $0.00023 | $0.00193 |
| Sonnet 5 | $0.00009 | $0.00077 |
| Haiku 4.5 | $0.00005 | $0.00039 |
Grade A, and why
agentos-draftsman 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 2d 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.
What it actually says
AgentOS Draftsman
Use this skill to turn approved design and engineering inputs into precise, reviewable, versioned documents. Verify tenant, project, site, domain, source-of-truth references, required format, audience, and current revision before drafting.
Workflow
collect approved source inputs
→ confirm document purpose and standard
→ draft with identifiers and revision metadata
→ run consistency and standards checks
→ route for Designer and Engineer review
→ incorporate tracked comments
→ issue approved revision
→ record distribution and supersede prior revision
Every document needs document ID, title, owner, project, source references, revision, status, author, reviewers, approver, issue date, and distribution scope. Never modify an approved revision in place; create a new revision and preserve the prior record.
Boundaries
Draftsman may format, diagram, revise, annotate, and prepare issue packages. Designer owns design intent, Engineer owns technical correctness, QA owns acceptance evidence, and Project Manager controls release. Require approval for controlled document issue, publication, supersession, external distribution, and changes that alter approved scope.
Quality checks
Check units, labels, references, legends, identifiers, revision history, page or canvas limits, accessibility, file integrity, and alignment with the approved source. Link comments to the exact revision and page or element. Do not introduce invented technical values.
Outputs
Return the draft or issued document reference, revision, source inputs, comments resolved, unresolved questions, reviewers, approval state, distribution list, evidence references, and next action. Keep controlled documents tenant-scoped and prevent obsolete revisions from being used for execution.
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.
- 2d ago First seen · 36 lines · 45 tokens per session scan A c265cb13a41e
agentos-draftsman is a skill published in the GitHub repository br3eze-code/br3eze-code (2 stars, last pushed 2d ago), licensed Apache-2.0. It adds 45 tokens to every session and 387 once invoked, about $0.0002 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 2026-08-31.
Other skills, from other repositories
anti-slop-frontend
A mechanical, countable anti-slop checklist for AI-generated frontend. Catches the specific signatures an undirected model defaults to: AI-purple glows, Inter-everywhere, em-dashes, div-based fake screenshots, eyebrow-on-every-section, beige+brass "premium" palettes, generic Jane Doe / Acme data. Advisory layer that…
frontend-mockup-loop-dashboard
Dashboard-specific adapter on the generic frontend-mockup-loop skill: binds the 7-step design loop to pi-agent-dashboard component sources, theme-system tokens, and isolated verification. Use when designing/redesigning any pi-agent-dashboard client surface. Triggers: "design a dashboard screen", "mockup a dashboard…
double-diamond
Double Diamond 디자인 방법론 — 문제 탐색과 해결책 개발.
using-designpowers
MUST run before any other Designpowers skill — shows welcome, checks taste profile, and routes to the correct first skill. Triggers on ANY design-related message. No other Designpowers skill may run until the welcome sequence has completed.
design-taste
Use when calibrating aesthetic direction — capturing design references, quality benchmarks, and the subjective qualities that make a design feel elevated. Invoked between strategy and design to give agents a shared sense of what "good" looks and feels like for this project.
add-branding
Customize app shell branding - themes, colors, navigation, logos. Works with app/brand/ and app/config/ declarative files.