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 vyuh-labs/dxkit --skill dxkit-author-extensiongit clone --depth 1 https://github.com/vyuh-labs/dxkitWrote 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/vyuh-labs/dxkit/dxkit-author-extension)<a href="https://agentmods.dev/skills/vyuh-labs/dxkit/dxkit-author-extension"><img src="https://agentmods.dev/badge/skills/vyuh-labs/dxkit/dxkit-author-extension/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/vyuh-labs/dxkit/dxkit-author-extension"><img src="https://agentmods.dev/badge/skills/vyuh-labs/dxkit/dxkit-author-extension.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 | $0.00113 | $0.01242 |
| Opus 5 | $0.00056 | $0.00621 |
| Sonnet 5 | $0.00023 | $0.00248 |
| Haiku 4.5 | $0.00011 | $0.00124 |
Grade A, and why
dxkit-author-extension 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 10d 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 — 89 lines — stays where its author put it; the contents beside it link to each section on GitHub.
dxkit-author-extension
You are writing the extension; the user describes the convention. Work the ladder from the bottom — the lowest rung that expresses the need wins, and every rung down is less code the user owns:
- Config / declared artifact — if the evidence already exists as a
Postman collection, Pact,
.httpfile, HAR, or OpenAPI doc, declare it in.dxkit/policy.json:flow.sourcesand STOP. No code. - External script (rung 3) — the user has extraction logic (any language) or you can write ~50 lines of Python: a manifest + stdin-payload adapter emitting one wire document. This covers almost everything: inventories, custom findings, delivery sinks.
- TypeScript plugin (rung 4) — only when the need is INSIDE dxkit's
gather: a bespoke HTTP-client wrapper flow can't see, a custom artifact
format for
flow.sources, base-URL logic beyondstripUrlPrefixes, or an assertion over the gathered flow model.
The authoring loop (all rungs)
- Scaffold:
vyuh-dxkit extensions init <name> --kind <kind> --command "…"(rung 3,--stubfor a Python starter) orvyuh-dxkit extensions init <name> --plugin [--kind <kind>](rung 4). - Read sample source/artifacts the user points at; fill in the logic.
vyuh-dxkit extensions dev <name>— field-precise errors ARE the spec; iterate until VALID and the summary shows the expected counts.- Sanity-check counts against ground truth (grep the repo, count rows).
- Commit the manifest + script/plugin + refreshed snapshot. Gates read the
committed snapshot offline;
refresh: on-mergekeeps it current via the extensions-refresh workflow.
Rung-4 contribution points (what goes in plugin.js)
The module's export is a DxkitExtensionDefinition (plain CommonJS object,
or defineExtension({...}) from @vyuhlabs/dxkit-sdk in TypeScript — it
stamps sdkMajor). Each key registers into an existing dxkit registry:
| Key | Use for | Example trigger phrase |
|---|---|---|
httpFlowDialect |
teach flow a bespoke client wrapper / niche framework — a data TABLE, no AST code | "our API calls go through acmeApi.request" |
contractReader |
a custom artifact format for flow.sources (kind, sniff, parse → raw calls/routes) |
"our contract fixtures are CSV files" |
urlNormalizer |
rewrite raw URLs ahead of canonical normalization ((url) => string | null, null = no opinion) |
"our clients call internal://svc/..." |
findingProducer / inventoryProducer / contractProducer / exporter |
the rung-3 wire protocol, in-process (manifest needs contributes + refresh + output) |
"we'd rather write TS than Python" |
integrationVerifier |
assert over the gathered flow model (ctx.flow.unserved = calls nothing serves); findings gate via the committed snapshot |
"fail the PR when a call has no backend route" |
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.
- 10d ago First seen · 89 lines · 113 tokens per session scan A c372d3e7e19a
dxkit-author-extension is a skill published in the GitHub repository vyuh-labs/dxkit (10 stars, last pushed 12d ago), licensed MIT. It adds 113 tokens to every session and 1,242 once invoked, about $0.0006 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 2026-08-31.
Other skills, from other repositories
csharp-nullable-reference-types
Guidelines for introducing and using nullable reference types (NRT) and System.Diagnostics.CodeAnalysis nullable attributes in C# / .NET codebases. Covers the nullability model, flow analysis, the null-forgiving operator, API design rules, the full attribute catalog (AllowNull, DisallowNull, MaybeNull, NotNull…
typescript-rules
React/TypeScript frontend development rules including type safety, component design, state management, and error handling. Use when implementing React components, TypeScript code, or frontend features.
ESLint for Test Quality
Enforce test quality with ESLint - eslint-plugin-jest, eslint-plugin-playwright, and eslint-plugin-testing-library rules in flat config, blocking focused tests, missing assertions, and flaky waits via a CI lint gate.
release-habit-hooks
Cut a new release of the habit-hooks packages. Use when asked to release, publish, or bump the version. Reviews what lands, enforces the in-sync versioning rule, validates the changelog, and drives the tag-triggered PyPI publish.
ring:adopting-lib-commons-huma-wrapper
Adopting the lib-commons/v5 shared Huma (OAS 3.1) OpenAPI wrapper + RFC 9457 problem model (commons/net/http/{openapi,problem}) in a Lerian Go service: wire openapi.New/ServeSpec + problem.Install (central >=500 scrub) on BOTH runtime and spec-gen paths, the per-rail problem.MapError flex seam, and rename-only spec…
implement-type-annotations
Add comprehensive type hints to Python/TypeScript code to improve IDE support, catch errors early, and enable better AI code understanding.