agentleFS
jqaisystemsjqaisystems/jqai-ai-skills/skills/research-to-article/SKILL.md

research-to-article

Skill3 starsChanged 6 months agoMITView on GitHub

Turn a topic and public source set into a source-traceable article package with claim mapping, metadata, links, a finished draft, and verification notes. Use when a user asks for a researched article, educational post, analysis, comparison, explainer, source-backed commentary, SEO or answer-focused article, or a publication-ready draft rather than a research brief.

The file

We split this skill into the rules it gives an agent

11 rules. About 0 are common, 9 are distinctive, and 2 are very specific to this project.

  • ≈0

    Common · estimate

    Many files like this have a rule of this kind.

  • ≈9

    Distinctive · estimate

    This author's own choices. Worth a look.

  • 2

    Very specific

    Tied to their project's names, files or tools. Probably skip.

  • Common
  • Distinctive
  • Very specific
  1. Very specific:

    Use research-brief-curator instead when the user wants only a digest, reading queue, or research brief.

    line 15
  2. Distinctive:

    Read references/source-quality.md before selecting evidence.

    line 21
  3. Distinctive:

    Prefer primary sources, official documentation, original research, standards, transcripts, and direct creator material.

    line 22
  4. Distinctive:

    Keep conflicting facts visible until a stronger source or explicit qualification resolves them.

    line 25
  5. Distinctive:

    Ask for a material editorial decision only when the evidence supports multiple incompatible directions.

    line 33
  6. Distinctive:

    Put the useful answer early and explain why the evidence matters.

    line 37
  7. Distinctive:

    Keep quotations short, necessary, and attributed.

    line 40
  8. Distinctive:

    Do not invent facts, quotes, statistics, examples presented as real, source access, product behavior, consensus, or outcomes.

    line 41
  9. Distinctive:

    Do not turn a single example, testimonial, or small study into a universal claim.

    line 42
  10. Very specific:

    Keep article publication and editorial update dates separate from source/event/video dates.

    line 47
  11. Distinctive:

    Check that explanatory images add information.

    line 48

≈ “Common” is an estimate: at least 1 in 5 of the 364,980 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.

Keep what fits and drop the rest. Your version, ready for your agents.
Files
SKILL.md

Research to Article

Build a finished article from current evidence while keeping facts, analysis, recommendations, and unresolved claims visibly separate.

1. Define the article contract

  • Confirm the topic, audience, practical purpose, time window, desired length, publication type, and required point of view.
  • Identify the intended output: editorial package, full draft, revision, or verification pass.
  • For video-led articles, identify the primary embed separately from supplemental sources and confirm the editorial date policy.
  • Use research-brief-curator instead when the user wants only a digest, reading queue, or research brief.
  • Preserve an approved angle unless new evidence makes it inaccurate or misleading.

2. Build the source and claim map

  • Browse current public sources whenever the topic includes facts that may have changed.
  • Read references/source-quality.md before selecting evidence.
  • Prefer primary sources, official documentation, original research, standards, transcripts, and direct creator material.
  • Record every material claim with its source, publication date, event date when different, evidence role, and limitations.
  • Separate sourced fact from analysis, inference, scenario, and recommendation.
  • Keep conflicting facts visible until a stronger source or explicit qualification resolves them.

3. Prepare the editorial package

  • Use assets/article-package-template.md.
  • Define the title, practical angle, slug, meta description, audience need, outline, internal-link opportunities, call to action, source credits, and Link Map.
  • For embeds or visual assets, read media and date handling. Record creator credit, source URL, verified source date, purpose, and permission or usage uncertainty.
  • Include only sections supported by the source map or clearly labeled as analysis.
  • Ask for a material editorial decision only when the evidence supports multiple incompatible directions.

4. Draft with evidence discipline

  • Put the useful answer early and explain why the evidence matters.
  • Cite material claims close to the wording they support.
  • Attribute preview, reported, estimated, disputed, or creator-tested information precisely.
  • Keep quotations short, necessary, and attributed. Prefer paraphrase with a direct source link.
  • Do not invent facts, quotes, statistics, examples presented as real, source access, product behavior, consensus, or outcomes.
  • Do not turn a single example, testimonial, or small study into a universal claim.

5. Verify the article package

  • Recheck names, dates, links, numerical claims, quotations, credits, title, slug, meta description, and Link Map.
  • Keep article publication and editorial update dates separate from source/event/video dates. Never backdate to a video date unless an explicit editorial policy requires that treatment, and flag misleading date labels.
  • Check that explanatory images add information; avoid repeating a video's player thumbnail while retaining separate social-preview assets.
  • Mark claims that remain qualified, inference, unsupported, or stale and revise or hold them.
  • Confirm the final draft distinguishes fact, analysis, and recommendation.
  • Report the evidence cutoff date and any source that could not be accessed.

6. Respect the publication boundary

Return the article package and verification state. Do not publish, modify a website, submit indexing requests, or update an external system. If the user later authorizes website implementation, hand the finished package to the appropriate website workflow and verify the affected surfaces with website-change-verifier.

Download now (ZIP)

Does it work?

Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.

No reviews yet

People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.

  • It worked
  • Partly
  • It didn't
What did you use it for? e.g. Reviewing pull requests in a Rails app

Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.

Your trail

Files you open will collect here, so you can find them again.

Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.

Now do this for your own.

Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.