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 aziontech/webkit --skill webkit-errorsgit clone --depth 1 https://github.com/aziontech/webkitWrote 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/aziontech/webkit/webkit-errors)<a href="https://agentmods.dev/skills/aziontech/webkit/webkit-errors"><img src="https://agentmods.dev/badge/skills/aziontech/webkit/webkit-errors/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/aziontech/webkit/webkit-errors"><img src="https://agentmods.dev/badge/skills/aziontech/webkit/webkit-errors.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 | $0.00152 | $0.04694 |
| Opus 5.5 | $0.00061 | $0.01878 |
| Sonnet 5.5 | $0.00030 | $0.00939 |
| Haiku 4.5 | $0.00015 | $0.00469 |
Grade A, and why
webkit-errors 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 — 329 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Skill: webkit-errors
Purpose
Every screen that talks to a server can fail, and almost every failure gets put in the wrong place — usually a toast, because a toast is the easiest thing to reach for and it always compiles. A toast is wrong for a field the user must fix (it dismisses itself and points nowhere), wrong for a sign-in (it flies away from the only object on the page and stacks one card per attempt), and right for exactly one case most apps never handle: the failure that arrives after the user has walked away.
This skill fixes where an error goes, what it carries when it gets there, and what it
says. It is the error half of webkit-ui-states (which owns the full loading / empty / partial
state surface) and the failure half of webkit-form (which owns the field and submit mechanics).
How to use
/webkit-errorsApply the placement table below to every failure path in this conversation./webkit-errors <file>Review the file and output, per gap:- the exact line / handler (quoted),
- which row of the table it breaks (1 short sentence),
- the concrete fix, naming the surface and what it must carry.
When to invoke
- Writing or reviewing any
catcharound a submit, a create, a delete, or a sign-in. - Building a signed-out screen (sign in, sign up, reset password, MFA).
- Building a long-running action (deploy, import, provision) the user can navigate away from.
- The user asks "where does this error go", "should this be a toast", "the error disappeared before I read it", "the user can't find what failed", "how do I show a 500".
How to find the components
Never guess an import path — files move. Resolve every primitive the same way:
- Ask the webkit MCP —
suggest_componentin plain words ("inline message", "toast", "password field"). - Or read
node_modules/@aziontech/webkit/catalog.json— every key underimportsis a real published subpath. If a subpath is not a key there, it does not exist.
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 First seen · 329 lines · 152 tokens per session scan A c503252d4522
webkit-errors is a skill published in the GitHub repository aziontech/webkit (3 stars, last pushed today), licensed MIT. It adds 152 tokens to every session and 4,694 once invoked, about $0.0006 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
sivtr-memory
Retrieve shared local work memory: terminal activity, AI conversation history, prior decisions, validation evidence, debugging trails, recaps, and handoff context. Use before asking the user to repeat local work context.
posthog
Session replays, product events, error tracking, feature flags and HogQL over a connected PostHog project. Use when the user asks what a user did before a bug, for a session recording or replay, for product analytics, for who an exception hit, or about a feature flag's rollout.
investigate-cloudflare
Investigate deployed Worker failures, regressions, and missing operational signals using Cloudflare Issues, logs, traces, SQL, or Observability MCP. Produce bounded evidence and verify recurrence against the deployed version.
debug
Structured bug diagnosis and fixing workflow that reproduces, diagnoses root cause, applies a minimal fix, writes regression tests, and scans for similar patterns.
sentry
Query Sentry organizations, projects, and unresolved issues/errors via the Sentry API. Use when the user asks about Sentry errors, issues, or projects.
signoz
Query observability (services, traces, logs, metrics) from a SigNoz instance via its API. Use when the user asks about app performance, errors, latency, or telemetry in SigNoz.