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-remediategit 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-remediate)<a href="https://agentmods.dev/skills/vyuh-labs/dxkit/dxkit-remediate"><img src="https://agentmods.dev/badge/skills/vyuh-labs/dxkit/dxkit-remediate/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-remediate"><img src="https://agentmods.dev/badge/skills/vyuh-labs/dxkit/dxkit-remediate.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 | $0.00000 | $0.02436 |
| Opus 5 | $0.00000 | $0.01218 |
| Sonnet 5 | $0.00000 | $0.00487 |
| Haiku 4.5 | $0.00000 | $0.00244 |
Grade A, and why
dxkit-remediate 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 9d 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 — 175 lines — stays where its author put it; the contents beside it link to each section on GitHub.
dxkit remediate — the agentic remediation lane
The lane plans the repo's debt (the grandfathered build/tests, advisories a version bump cannot fix, the lint backlog) as finite work orders and works them in two tiers: deterministic recipes execute first at $0, and only the remaining orders go to a scoped agent, one order per run under a derived budget and an enforced envelope. Everything lands INSIDE dxkit's verified frame: an entry-attributed correctness floor and the guardrail run before any PR opens, and the agent's own claim of success is never trusted. One standing PR per task; the PR body is the verification ledger plus the agent envelope (model, turns, spend, outcome).
Preview before spending anything
vyuh-dxkit remediate plan
Shows, per configured task, the full model resolution chain (task → tier → driver-native model), the budget caps and which ones the driver can enforce, and whether the agent CLI resolves here. It also lists the planned WORK ORDERS: the finite units the run would actually work (class, tier, derived budget, done criterion), the scheduled matrix derived from them (a task with no open orders spawns no scheduled job), any classes the circuit breaker has paused (with the reason and the unpause conditions), and the spend-ceiling trim. No key needed, no spend.
Model settings (remediate.agent.model):
auto(default) — each task uses its registry tier: light for mechanical work (fix-lint), standard for reasoning work (fix-build, fix-vulns, improve-tests). Tiers map to the driver's rolling aliases, so they never break when a model generation rolls over.- a tier name (
light|standard|deep) — pins all tasks; portable across drivers.deepis the most capable and most expensive; nothing selects it automatically. - anything else — a driver-native model id passed through verbatim (a driver switch may invalidate it; a dated id will eventually deprecate).
Run one task locally (human present)
vyuh-dxkit remediate --task fix-lint # run + verify, no PR
vyuh-dxkit remediate --task fix-vulns --land pr # land the standing PR
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.
- 9d ago First seen · 175 lines · 0 tokens per session scan A 775885b4341e
dxkit-remediate is a skill published in the GitHub repository vyuh-labs/dxkit (10 stars, last pushed 11d ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 2,436 tokens. 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
USMetrics
Analyze and update 68 US economic and social indicators from five government APIs (FRED, EIA, Treasury, BLS, Census) across 10 categories: GDP, Inflation, Employment, Housing, Consumer Finance, Markets, Trade, Government/Fiscal, Demographics, Health. Two workflows: UpdateData, GetCurrentState (10y/5y/2y/1y trend…
company-cfo
Monthly CFO workflow for a company or agency — pull raw data from bank + payment processor + payroll + expense management, categorize and reconcile, compute end-of-month cash via transaction-sum method, update a scenario projector for forward forecasting, write the monthly snapshot report, surface decisions to…
personal-cfo
When you want to model personal financial scenarios — house purchase + rental income (ADU, bedroom rentals, house-hacking), renovation budgets, monthly cash flow forecasts, big-purchase decisions, savings/investment what-ifs. For personal life: a household (you + partner), household budgets, real-estate decisions.…
integration-e2e-testing
Integration and E2E test design principles, ROI calculation, test skeleton specification, and review criteria. Use when designing integration tests, E2E tests, or reviewing test quality.
recipe-front-review
Reviews completed frontend implementation for governing-source compliance, scope economy, repository quality, and security, then applies user-approved React corrections.
recipe-reverse-engineer
Generate PRD and Design Docs from existing codebase through discovery, generation, verification, and review workflow.