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 kensaurus/cursor-kenji --skill enhance-web-formsgit clone --depth 1 https://github.com/kensaurus/cursor-kenjiWrote 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/kensaurus/cursor-kenji/enhance-web-forms)<a href="https://agentmods.dev/skills/kensaurus/cursor-kenji/enhance-web-forms"><img src="https://agentmods.dev/badge/skills/kensaurus/cursor-kenji/enhance-web-forms/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/kensaurus/cursor-kenji/enhance-web-forms"><img src="https://agentmods.dev/badge/skills/kensaurus/cursor-kenji/enhance-web-forms.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 | $0.00059 | $0.01729 |
| Opus 5.5 | $0.00024 | $0.00692 |
| Sonnet 5.5 | $0.00012 | $0.00346 |
| Haiku 4.5 | $0.00006 | $0.00173 |
Grade A, and why
enhance-web-forms 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 — 155 lines — stays where its author put it; the contents beside it link to each section on GitHub.
enhance-web-forms — Production-Quality Forms
Degree of freedom: MIXED. Schema and state-machine judgment [HIGH freedom]; a11y probes and playwright walks [LOW freedom — run exactly].
Forms are where users hand you their data and where apps most often feel broken: unlabeled fields, validation that fires on every keystroke, errors screen readers never announce, a submit button that does nothing visible for three seconds, and no recovery when the request fails. This skill fixes all of that on real forms in the repo.
A form is done when it is accessible, validated on both sides, gives feedback for every state, and recovers from failure. Compile-clean is not done.
Before any browser interaction, read protocol-browser-anti-stall and apply it.
How to reason
- Inventory — forms, library, schema; bow out if none
- Structure — programmatic labels, types, error association
- Parity — one schema both sides; blur + submit, not every keystroke
- States — submitting / success / error / unsaved; recover on failure
Worked example
Inventory: signup uses raw
<form>+ inlineif (!email); API validates with zod. Structure:htmlForlabels;autocomplete="email";aria-invalid+ summary. Parity: share the zod schema; validate on blur/submit; map "email taken" to the field. States: submit disables +aria-busy; values preserved on 500; dirty-nav warn.
Self-critique before reporting
- Labeled — placeholder is never the only name
- Both sides — client and server rules match; error shapes map to fields
- Walked — empty, invalid, valid, and server-error paths in the browser
- Right owner — page/flow UX beyond the form →
enhance-web-ux; WCAG sweep →audit-accessibility
Phase 0 — Detect the stack and inventory forms [HIGH freedom]
cat package.json | grep -iE "react-hook-form|formik|@tanstack/react-form|final-form|zod|yup|valibot|superstruct"
rg -n "<form|onSubmit|useForm|handleSubmit" -g "*.{tsx,jsx,vue,svelte}" -l
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 · -3 tokens per session bfa2b27866b3
- 1mo ago First seen · 155 lines · 62 tokens per session scan A 7a5bac3c4d57
enhance-web-forms is a skill published in the GitHub repository kensaurus/cursor-kenji (9 stars, last pushed 26d ago), licensed MIT. It adds 59 tokens to every session and 1,729 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 2026-09-03.
Other skills, from other repositories
ai-evals-builder
Build AI evals using the Husain-Shankar framework (error analysis, open/axial coding, LLM-as-judge). Use when a user needs to create, improve, or debug evals for an AI product — including defining failure modes, building LLM judges, or setting up production monitoring for an LLM application.
test-driven-development
Use when implementing any feature, bugfix, or code change - write tests first, watch them fail, then implement.
patrol-e2e-testing
Use when writing E2E/integration tests, testing native interactions like permissions or system dialogs, capturing UI regressions, or validating cross-platform behavior (Patrol 4.x).
mocktail
Use when creating mocks, stubbing methods, verifying interactions, registering fallback values, or choosing between mocks, fakes, and real objects (Mocktail).
instruction-ablation
Rebuild a minimal instruction set from evidence: confirm a bare baseline, run one real task in-session under guardrails only, log struggles, and add an instruction only after the same failure repeats; retest and keep only if verification improves. Use when the user invokes /instruction-ablation or asks to ablate…
component-creation
Step-by-step workflow for creating accessible, tested UI components. Use when the user asks to create a new UI component.