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 tomzx/agents --skill research-articlegit clone --depth 1 https://github.com/tomzx/agentsWrote 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/tomzx/agents/research-article)<a href="https://agentmods.dev/skills/tomzx/agents/research-article"><img src="https://agentmods.dev/badge/skills/tomzx/agents/research-article/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/tomzx/agents/research-article"><img src="https://agentmods.dev/badge/skills/tomzx/agents/research-article.svg" alt="Reviewed on agentmods" width="80" height="20"></a>- NVIDIA SkillSpector warn
SkillSpector: 1 finding, up to medium
These are SkillSpector’s own severities. On a checked sample its high-severity flags on skills were ~96% false positives — a documented command, a public API, a “never do X” rule — so we show them as a caution to read, not a verdict. Why →
- medium Excessive Agency · line 69 Skill enables autonomous high-impact decisions without human-in-the-loop verification. Critical operations (destructive commands, financial transactions, data deletion) should require explicit user confirmation.Fix: Add human-in-the-loop confirmation for destructive, irreversible, or high-impact operations. Never auto-execute commands that modify files, send data, or alter system state.
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 | $0.00047 | $0.01870 |
| Opus 5 | $0.00023 | $0.00935 |
| Sonnet 5 | $0.00009 | $0.00374 |
| Haiku 4.5 | $0.00005 | $0.00187 |
Grade A, and why
research-article 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.
How it starts
The opening of the file, as written. The whole thing — 185 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Research Article
Given a topic, discovers what the state of the art is by searching broadly, triaging sources, deep-reading the strongest ones, and synthesizing the landscape.
The output is a research brief that becomes the input to /create-article, so the writing step can focus on craft rather than discovery.
Prerequisites
- Topic: The subject to investigate (e.g., "vector databases", "test-time compute in LLMs", "CRDTs for collaborative editing")
- Scope (optional but recommended): The specific angle or question, time bounds (e.g., "last 2 years"), and what "state of the art" means here (best accuracy, fastest, most adopted, most recent breakthrough)
- Intended audience (optional): Helps calibrate which sources and framing are relevant
- Seed sources (optional): Any URLs, papers, or files the user already knows matter
Steps
- Clarify the topic and scope. State the specific research question in one sentence. Define what "state of the art" means for this topic (top benchmark score, dominant method, industry standard, latest breakthrough). Note any time bounds or constraints.
- Cast a wide net. Search multiple source types (see Search Strategy below). Start with surveys, review papers, and benchmark leaderboards to map the terrain quickly before going deep.
- Build a candidate source list. For each result, capture title, URL, type, year, and a one-line relevance note. Prioritize primary sources (papers, official docs, benchmarks) over secondary commentary, but keep influential blog posts and talks that shape practice.
- Triage. Rank candidates by relevance, credibility, and recency. Keep the strongest 8-15 to deep-read. Drop outdated or low-quality sources, but record why so the exclusion is deliberate.
- Deep-read the top sources. Fetch and read each one. Extract: the problem it tackles, its approach, its key results and numbers, and how it compares to prior work. Verify claims against the source rather than paraphrasing from memory or a summary.
- Map the landscape. Group the sources into the main approaches or camps. Identify what currently defines the state of the art (top scores, dominant technique, widely adopted standard). Note where the field agrees and where it actively disagrees.
- Surface gaps and open questions. What is unsolved, contested, or just emerging? These tensions are what make an article worth reading.
- Synthesize. Write the state-of-the-art summary. Recommend an angle for the article: the single core insight or argument the research best supports.
- Produce the research brief using the output format below.
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.
- 6d ago First seen · 185 lines · 47 tokens per session scan A e1e90f41e2f9
research-article is a skill published in the GitHub repository tomzx/agents (6 stars, last pushed yesterday), licensed MIT. It adds 47 tokens to every session and 1,870 once invoked, about $0.0002 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 2026-09-03.
Other skills, from other repositories
systematic-debugging
Use when encountering any bug, test failure, or unexpected behavior, before proposing fixes.
local-ai-agents
Build local-first AI agents that run entirely on a developer workstation with Microsoft Foundry Local and Qwen function-calling models. Covers Small Language Models (SLMs), the OpenAI-compatible local endpoint, sandboxed local tools, local RAG with Chroma, local MCP servers, hybrid cloud/local routing, and the…
next-cache-components-adoption
Turn on Cache Components in a Next.js app and resolve the blocking routes it surfaces. Use when the user wants to enable, adopt, or migrate to Cache Components, flip the cacheComponents flag, work through a flood of blocking-prerender / instant validation errors, run the cache-components-instant-false codemod, or…
insight-error-page
Write or audit an insight-kind error page for the Next.js dev overlay. Use when creating a new errors/ .mdx page, auditing an existing one, or checking that a page matches the framework fix cards. Covers page structure, title alignment, FixCard cards with Copy prompt button, code snippets, terminology verification…
next-cache-components-optimizer
Drive a Next.js route to instant navigation by setting up an agentic loop, under Cache Components / PPR, on initial load (hard navigation) and client-side navigation (soft navigation). Encode the goal as a failing @next/playwright instant() e2e and work it to green, one verified route at a time; the shipped test then…
next-partial-prefetching-adoption
Turn on Partial Prefetching in a Next.js app and work through the insights it surfaces. Use when the user wants to enable or adopt Partial Prefetching, flip the partialPrefetching flag, opt routes in with export const prefetch = 'partial', audit Link prefetch={true} behavior, preserve existing prefetched UI with…