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 alex-durango/pingfusi --skill fix-with-pingfusigit clone --depth 1 https://github.com/alex-durango/pingfusiWrote 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/alex-durango/pingfusi/fix-with-pingfusi)<a href="https://agentmods.dev/skills/alex-durango/pingfusi/fix-with-pingfusi"><img src="https://agentmods.dev/badge/skills/alex-durango/pingfusi/fix-with-pingfusi/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/alex-durango/pingfusi/fix-with-pingfusi"><img src="https://agentmods.dev/badge/skills/alex-durango/pingfusi/fix-with-pingfusi.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector pass
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 | $0.00078 | $0.01132 |
| Opus 5 | $0.00039 | $0.00566 |
| Sonnet 5 | $0.00016 | $0.00226 |
| Haiku 4.5 | $0.00008 | $0.00113 |
Grade A, and why
fix-with-pingfusi 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 9d 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 — 69 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Fix a draft with pingfusi (the review loop)
The draft is the user's CURRENT project (or a directory/URL they name). Any builder's output works — pingfusi runs the loop that closes the last 10%: the draft is reviewed side-by-side with the original, pins come back on what's off, you fix them in the draft's own source, repeat until an approving verdict.
Steps
-
Preflight:
pingfusi doctor(surface failures). Identify the ORIGINAL url — ask the user if it isn't obvious from the project (README, comments, git remote). -
Build the draft: run its production build and identify a self-contained output directory (
dist/,build/,out/, or equivalent). A running dev server is useful for editing, but it is not the preferred reviewer URL. -
Register:
pingfusi adopt <name> <original-url>(name = short slug). This is the reviewer-loop-only path — no pixel gates, the review verdict is the check. -
Publish through Pingfusi hosting by default:
pingfusi publish <built-dir> --target <name> --name <name>-reviewThis records the immutable hosted URL where
pingfusi reviewalready looks for it. Rebuild and publish again after each change. Usepingfusi tunnel <name> --url http://localhost:<port>only when the production app genuinely requires a live server (SSR, server actions, API routes, or authentication) and cannot produce a self-contained build. Never choose a tunnel merely because a dev server is already running. -
The loop:
pingfusi review <name> file [--region "…"] [--context "one line: what this site/page is"] [--results 1..20]→ tell the user the round is filed with an independent human reviewer on the pingfusi service (the reviewer pins what's wrong + picks a verdict — the user does not review). The filing command automatically chains client-safe wait legs until feedback. If a raw MCP leg returns pending, immediately callpingfusi_waitagain; never return pending to the user or file a duplicate. Passive result/verify reads do not renew the lease (a lapse only pulls the round from the feed for new claims; a reviewer mid-review can still finish). On each verdict:- Approved → done; report with the round history.
- Pins → fix each in the DRAFT'S OWN source (its components/styles — match its
idioms; derive fixes from the original site's real markup/CSS, never invent),
rebuild and publish a new immutable hosted draft (or, for the exceptional live-server
path, run
pingfusi tunnel <name> --check), refile with--changelog "what changed since your last review"; that refile owns its wait. Repeat — the run is not complete until an approving verdict is recorded. Full rounds default to 1 result; request 5 for a broader read and 15–20 only for complex work or when higher confidence is worth it. Each completed result costs 1 credit; undelivered results are not charged. - If a pin is temporal (timing, easing, spring, stagger, scroll/pointer-driven,
canvas, or WebGL motion), run
pingfusi next <name>and follow itspingfusi motion …machine route (first-draft doctrine: the draft build reproduces animations automatically; motion checks are build receipts and warnings — never gates, never review rounds). Fix with the routed utility —motion passto re-run the build pass,verify-introspectedfor the exact engine-declaration diff, or the sampled chainsample→apply-sampled→verify-sampled— then redeploy and refile the page round. Motion never blocks filing or refiling; the reviewer's side-by-side look IS the motion review.
-
Rules: all review contact through
pingfusi review <name> …(never any MCP directly); polls one-sided only — a poll is text-only and the reviewer sees no page, so it may only ask what the ORIGINAL site does, described in words; never ask a poll to compare, judge, or look at anything (that is the review round's job); never submit or open a review yourself; screenshots for triage, never as proof of a match. No login? Stop and have the user runpingfusi setup— review rounds require it; there is no offline review path. If anything else blocks a round, stop and tell the user what failed.
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.
- 9d ago First seen · 69 lines · 78 tokens per session scan A 56dde921d725
fix-with-pingfusi is a skill published in the GitHub repository alex-durango/pingfusi (110 stars, last pushed 5d ago), licensed MIT. It adds 78 tokens to every session and 1,132 once invoked, about $0.0004 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
web-prototype
General-purpose desktop web prototype. Single self-contained HTML file built by copying the seed assets/template.html and pasting section layouts from references/layouts.md. Default for any landing / marketing / docs / SaaS page when no more specific skill matches.
web-prototype-taste-brutalist
Swiss industrial-print web prototype. Newsprint canvas, monolithic black grotesque, viewport-bleeding numerals, hairline grid dividers, hazard-red accent, ASCII syntax decoration. Distilled from Leonxlnx/taste-skill brutalist-skill (Swiss Industrial Print mode).
frontend-design
Guidance for distinctive, intentional visual design when building new UI or reshaping an existing one. Use whenever the task produces or modifies anything a user will see rendered — websites, landing pages, web apps, dashboards, React/HTML/Vue components, artifacts with visual output, style overhauls, or "make this…
frontend-design
Produce intentional, responsive, accessible UI work and perform visual QA instead of generic component assembly.
image-to-code
Implement a selected image, screenshot, mockup, generated concept, or design frame as a faithful, responsive, interactive frontend in the target codebase.
frontend
MUST USE for frontend/web UI/UX/visual work: building, styling, redesigning pages/components, React setup, performance audits, visual QA, taste, and polish. Routes four rulesets: design taste router and brand references; perfection for Playwright/Chromium Lighthouse/Core Web Vitals; ui-ux-db palettes/fonts/guidelines…