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 naruto1031/uproad-plugin --skill reviewgit clone --depth 1 https://github.com/naruto1031/uproad-pluginWrote 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/naruto1031/uproad-plugin/review)<a href="https://agentmods.dev/skills/naruto1031/uproad-plugin/review"><img src="https://agentmods.dev/badge/skills/naruto1031/uproad-plugin/review/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/naruto1031/uproad-plugin/review"><img src="https://agentmods.dev/badge/skills/naruto1031/uproad-plugin/review.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 | $0.00064 | $0.01247 |
| Opus 5.5 | $0.00026 | $0.00499 |
| Sonnet 5.5 | $0.00013 | $0.00249 |
| Haiku 4.5 | $0.00006 | $0.00125 |
Grade A, and why
review 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 27d 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 — 71 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Close one review round
The client left comments on a link you sent. This skill turns that into a new version they can look at, and leaves an honest record of what was and was not done.
There is one hard stop: the fixes are shown to you locally before they replace what the client sees. The link is already in their inbox, so a wrong fix reaches them immediately.
1. Read what came back
list_designs— find the design if the user did not name one. Match on title orexternal_key. Every row carries itsworkspace; the list spans every workspace the connection can reach, so if the same title or key turns up in two workspaces, ask which one rather than picking. Everything after this is keyed bydesign_id, which already pins the workspace.list_commentswithunresolved_only: true.
If nothing is unresolved, say so and stop. Do not invent work.
Read all of them before touching anything. Clients contradict themselves across comments, and the second one usually wins — better to notice that now than to implement both.
Sort each comment into one of three piles, and say which pile you put things in:
- Change requests — do them
- Questions — the client is asking, not instructing. You cannot answer these: the MCP tools have no reply. Collect them and hand them to the user to answer in the app
- Out of scope or contradictory — anything that needs a decision you should not make alone (budget, scope, something that breaks a stated requirement). Surface it, do not quietly implement or quietly drop it
Comments carry pin_x/pin_y when the client pinned them to a spot on the prototype. Use that to locate what they meant, and say which element you think a pin refers to when it is ambiguous — a misread pin produces a confidently wrong fix.
2. Make the changes
get_design_html— always start from what is actually deployed, not from whatever is on disk. Someone else may have pushed since. Write it intodesigns/<slug>/index.html, which is where designs live (one directory each, matching/uproad:new). Derive<slug>from the design'sexternal_keywhen it has one.- Apply the changes. Keep the edits tight; unrelated refactoring makes the next round harder to review.
- If the feedback changes what the thing does rather than how it looks, the spec is now wrong too. Fetch it with
get_doc(spec.md) and updatedesigns/<slug>/spec.md. A spec that silently drifts from the prototype is worse than no spec — the next engineer trusts it. Do not push it yet; it goes out with the prototype in step 4.
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.
- 27d ago Changed a982b2604a27
- 29d ago Changed 42192c227332
- 1mo ago First seen · 71 lines · 64 tokens per session scan A e488f07fedde
review is a skill published in the GitHub repository naruto1031/uproad-plugin (0 stars, last pushed 28d ago), licensed MIT. It adds 64 tokens to every session and 1,247 once invoked, about $0.0003 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 2026-08-31.
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.