dxkit-gate

dxkit-gate is a skill for Claude Code from vyuh-labs/dxkit. It costs 94 tokens per session (745 once invoked), scanned A, original, MIT.

A checker for a directory containing generated code, an exported file tree, or several service folders. It compares the contents with a policy and returns a machine-readable verdict, without requiring Git.

In plain words
What is it for?
Use it to check generated packages, exported trees, edited directories, or service workspaces against a policy. Programs can read its `verdict.v1` JSON result.
Why use it?
It shows whether new or edited files meet the chosen rules, while separating newly introduced problems from issues that already existed. It can also check how services in a workspace connect and whether declared process steps are served.

Skill for Claude Code

Written for Claude Code: installed under .claude/.

Install

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.

agentmods
npx agentmods add skills/vyuh-labs/dxkit/dxkit-gate
Any agent
npx skills add vyuh-labs/dxkit --skill dxkit-gate
Clone the repo
git clone --depth 1 https://github.com/vyuh-labs/dxkit

Made for: Claude Code.

Wrote 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.

agentmods badge for dxkit-gate

README.md
[![agentmods](https://agentmods.dev/badge/skills/vyuh-labs/dxkit/dxkit-gate.svg)](https://agentmods.dev/skills/vyuh-labs/dxkit/dxkit-gate)
Your own site
<a href="https://agentmods.dev/skills/vyuh-labs/dxkit/dxkit-gate"><img src="https://agentmods.dev/badge/skills/vyuh-labs/dxkit/dxkit-gate.svg" alt="Measured on agentmods" height="20"></a>
Per session 94 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 745 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
Origin original No closer match found in the catalogue.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00094 $0.00745
Opus 5 $0.00047 $0.00373
Sonnet 5 $0.00019 $0.00149
Haiku 4.5 $0.00009 $0.00075

Measured 6d ago against content hash d5cd0c7aa02a, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-06, from the pricing page.

Security

Grade A, and why

dxkit-gate 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 6d 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.

src-templates/.claude/skills/dxkit-gate/SKILL.md · 60 lines

How it starts

The opening of the file, as written. The whole thing — 60 lines — stays where its author put it; the contents beside it link to each section on GitHub.

dxkit-gate

This skill runs the embeddable tree gate: one call in, one machine-readable verdict out. It is the same engine as the repo guardrail pointed at a bare directory, so the verdicts and fingerprints match what an onboarded repo would see.

Choosing the prior

  • Fresh tree (default): vyuh-dxkit gate <dir> --policy <policy.json>. Everything found is net-new by construction. Right for a freshly generated or converted package.
  • Edited tree: add --baseline <original-dir> to diff against the generated original. Only findings the edit introduced can block; pre-existing ones are grandfathered exactly like repo-mode debt.
  • Workspace / wave: vyuh-dxkit gate <dir> --workspace --flows <flowsDir> judges every immediate subdirectory as a member tree AND the composition: unresolved cross-member calls block, routes nobody consumes warn, and a declared flow (*.flow.json, flow.v1) with an unserved step blocks.

Always pass --json when a program (including you) reads the result; parse the verdict.v1 document, not the human text.

Trust: decide it deliberately

The gate treats the tree as untrusted by default: it scans bytes but never executes the tree's code, and the correctness floor (compile + tests) plus command checks are skipped with a disclosed cause. Pass --trusted ONLY when the user is prepared to execute the tree's own code. Ask if unclear; do not silently add --trusted to make skips go away.

Reading the verdict

  • Exit 0 passed / 1 blocked / 2 cannot_gate. Treat cannot_gate as a refusal with a named reason and remedy, never as a pass or a failure to fix in the code.
  • Each blocking finding carries a stable fingerprint, a file locator (member-prefixed in wave mode), and the check that produced it.
  • checks[] entries with a skippedWithCause are disclosures, not errors. Report them to the user (especially untrusted-skips) so "passed" is never read as broader coverage than actually ran.
  • The verdict names the policy (id, version, content hash). When comparing two verdicts, confirm the policy identity matches first.

Read the full file on GitHub · 60 lines

Changes

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.

  1. 6d ago First seen · 60 lines · 94 tokens per session scan A d5cd0c7aa02a

Subscribe to this mod's changes

dxkit-gate is a skill published in the GitHub repository vyuh-labs/dxkit (10 stars, last pushed 8d ago), licensed MIT. It adds 94 tokens to every session and 745 once invoked, about $0.0005 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.

Related

Other skills, from other repositories

habit-hooks-review

Spawn a reviewer sub-agent to assess a change set against habit-hooks's coding principles. Use AFTER habit-hooks reports clean — habit-hooks catches structural smells; this catches what it cannot (correctness, tests, design, missed edge cases).

habit-hooks/habit-hooks · 51 tokens

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.

habit-hooks/habit-hooks · 56 tokens

habit-hooks-prompting

Write or revise a habit-hooks coaching prompt. Use when a linter / knip / jscpd rule fires and the agent's default fix is wrong or shallow, or when adding a project-local override prompt. Keeps prompts short and outcome-focused using the ROSE pattern.

habit-hooks/habit-hooks · 61 tokens

brainstorming

You MUST use this before any creative work - creating features, building components, adding functionality, or modifying behavior. Explores user intent, requirements, and design before implementation.

HumanBean17/jrag · 38 tokens

writing-plans

Use when you have a spec or requirements for a multi-step task, before touching code.

HumanBean17/jrag · 21 tokens

explore-codebase

MUST BE USED PROACTIVELY. Universal read-only codebase exploration. Combines jrag graph navigation (call chains, routes, service boundaries, impact analysis, FQN resolution) with broad file-system search (grep, glob, file reading). Use for any exploration: locating code, tracing dependencies, finding patterns, 'where…

HumanBean17/jrag · 117 tokens