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 radimsem/forge-skills --skill blueprintgit clone --depth 1 https://github.com/radimsem/forge-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/radimsem/forge-skills/blueprint)<a href="https://agentmods.dev/skills/radimsem/forge-skills/blueprint"><img src="https://agentmods.dev/badge/skills/radimsem/forge-skills/blueprint/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/radimsem/forge-skills/blueprint"><img src="https://agentmods.dev/badge/skills/radimsem/forge-skills/blueprint.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 | $0.00062 | $0.01011 |
| Opus 5.5 | $0.00025 | $0.00404 |
| Sonnet 5.5 | $0.00012 | $0.00202 |
| Haiku 4.5 | $0.00006 | $0.00101 |
Grade A, and why
blueprint 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 — 108 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Blueprint
Explore UI options in the browser, approve a design, and hand off a spec.
Overview
Read the project's design language and show interactive proposals in the browser. The user pastes each screen's copied response into the agent session. That paste is the only return channel; the companion server keeps no session state. After approval, write the spec and offer a handoff to plans, tickets, or orchestration.
The interactive machinery ships in scripts/ and is never reimplemented inline; each session authors only HTML content fragments per references/screens.md.
Parameters
Flags compose and are parsed from anywhere in the invocation; the canonical matrix with composition rules is references/flags.md.
| Flag | Effect |
|---|---|
fresh |
Force a full design re-harvest, ignoring the .brainstorm/ cache |
terminal |
Skip the browser; run every round in the terminal fallback |
resume |
Continue from an existing ledger: restart the server, re-push the last unresolved screen |
When to Use
Use for UI/UX brainstorming: new screens, redesigns, component-level polish, anything where the user should see options before choosing. If the topic has no visual dimension (pure backend or API design), say so in one line and defer to a plain brainstorming workflow instead of forcing a browser on a text problem.
Workflow
Part 1 — Setup (1–3): classify → harvest → start companion
Part 2 — Rounds (4–5): push screen → read pasted response, ledger it (loop)
Part 3 — Close (6–8): [GATE] recap + approval → spec → handoff
Step 1: Classify & parse
Confirm the topic is visual (see When to Use). Parse flags. Enumerate the UI items in scope. Sessions typically batch several; each becomes a track through the rounds.
Step 2: Design harvest
Build or refresh .brainstorm/style.css and .brainstorm/design-notes.md
per references/harvest.md. No screen may be pushed
before the harvest exists.
What ships with it
10 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.
- references/anti-patterns.md 1.0 KB
- references/composer.md 2.2 KB
- references/flags.md 695 B
- references/handoff.md 2.5 KB
- references/harvest.md 2.1 KB
- references/ledger.md 1.5 KB
- references/screens.md 3.0 KB
- scripts/composer.js 7.3 KB runs code
- scripts/frame.html 2.4 KB
- scripts/serve.mjs 4.6 KB runs code
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 · -2 lines e6c92b9ea8b6
- 1mo ago First seen · 110 lines · 62 tokens per session scan A 5ea1e5a87dc3
blueprint is a skill published in the GitHub repository radimsem/forge-skills (15 stars, last pushed 1mo ago), licensed MIT. It adds 62 tokens to every session and 1,011 once invoked, about $0.0002 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 2026-08-30.
Other skills, from other repositories
webgl-holographic-foil
A self-contained WebGL2 hero: thin-film interference over a crushed-foil surface whose palette shifts with the viewing angle; move the cursor to tilt the film.
html-ppt-hermes-cyber-terminal
OpenDesign + BYOK: choosing and wiring your own model, hands-on — cost, quality, and the routing decision. Built as a decision-grade AI literacy deck for engineers, IT, applied-AI teams.
html-ppt-taste-brutalist
16:9 HTML deck in tactical-telemetry / CRT-terminal taste. Deactivated-CRT charcoal slides, white-phosphor monospace, hazard-red accent, scanline overlay, ASCII syntax, density over decoration. Distilled from Leonxlnx/taste-skill brutalist-skill (Tactical Telemetry mode).
visual-ralph
Visual Ralph orchestration for frontend UI from generated references, static references, or live URL targets, using $ultragoal with built-in visual verdict and pixel-diff evidence until the implementation matches and leaves a reproducible design system.
glance-canonical-layouts
Use this skill to create home screen widgets using Jetpack Glance Canonical Layouts from the Android Platform Samples project, or when asked to implement standard widget layout patterns (including Check List, Image Grid, Snap Scrolling, and Standard Toolbar).
accessibility
Consolidated accessibility skill entrypoint for WCAG 2.2, ARIA Authoring Practices, cognitive accessibility, Section 508, EN 301 549, design intent verification, and the Accessibility Planner workflow.