flaqai/Awesome_SEO_Writing_Skill / writer/medium-writer
medium-writer
VerifiedMedium-native long-form research, author-assistance, writing, editing, review, topic discovery, publication matching, and packaging workflow. Use when creating, outlining, researching, revising, auditing, or packaging Medium stories, essays, tutorials, reviews, explainers, comparisons, case studies, publication submissions, Medium 长文, Medium 选题, Medium Topics, or Medium publishing packs that need original author perspective, narrative craft, relevant topics, responsible AI disclosure, and reader-centered distribution readiness.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 21.9 KB |
| references/medium-article-template.md | markdown | 16.1 KB |
| references/medium-review-rubric.md | markdown | 8.7 KB |
| references/medium-topic-research.md | markdown | 17.7 KB |
| scripts/audit-medium-markdown.mjsexecutable | script | 13.5 KB |
references/medium-topic-research.md
View on GitHub# Medium Topic Research and Story Opportunity Map
Use this reference before outlining a Medium story, choosing Topics, or recommending a Publication. The goal is not to imitate what already performs. The goal is to find a live reader conversation where this author can add experience, evidence, craft, or a fresh synthesis.
## Research Principle
A strong Medium opportunity sits at the intersection of:
1. a reader curiosity, problem, emotion, or decision;
2. a visible Medium Topic or Publication audience;
3. an author advantage that cannot be produced by generic summarization;
4. enough evidence or narrative material to support a substantial story;
5. an angle that respects the reader rather than capturing traffic for its own sake.
Do not confuse Topic size, story volume, follower count, claps, recommendations, search indexing, and reader demand. Each is a different signal.
## Source Hierarchy
Use the strongest available sources in this order:
1. User-provided experience, research, notes, experiments, decisions, interviews, drafts, or audience feedback.
2. Primary sources: official documentation, original reports, research papers, standards, datasets, or direct statements.
3. Medium Topic pages, recent recommended stories, Staff Picks, and Publication archives.
4. Publication submission guidelines and About pages.
5. The author's Medium Stats, reader interests, responses, highlights, and prior-story patterns when available.
6. General web search, search trends, forums, and secondary analysis as discovery aids.
Search snippets and popular stories can reveal vocabulary or questions. They do not prove a claim and must not supply copied structure, phrasing, or personal material.
## Medium Search Procedure
When Medium is accessible:
1. Search the core subject as a phrase and as a natural-language question.
2. Open the closest established Topic page rather than accepting a misspelled or tiny Topic with a similar name.
3. Record adjacent Topics shown on that Topic page.
4. Review a small sample of recent recommended stories and older durable stories.
5. Record repeated title mechanisms, reader questions, story formats, and obvious content gaps.
6. Search for relevant Publications and inspect their recent output, About page, and submission guidelines.
7. Compare profile-first publishing with Publication submission.
8. Triangulate any current or factual premise with primary external evidence.
9. Identify what only this author can add.
Do not infer that Medium's recommended order is a stable ranking. Feeds and recommendations may be personalized and change over time.
## Query Grid
Generate 8-20 focused queries across these dimensions:
| Dimension | Examples |
|---|---|
| Core subject | AI agents, product design, grief, startup failure, causal inference |
| Reader | data scientist, founder, first-time manager, designer, caregiver |
| Tension | speed vs quality, ambition vs burnout, automation vs judgment |
| Experience | what I learned, field notes, experiment, failure, migration, recovery |
| Method | tutorial, teardown, framework, annotated example, build log |
| Change | 2026 shift, new policy, after AI adoption, post-launch lesson |
| Misconception | what everyone repeats, advice that failed, metric that misleads |
| Consequence | career, identity, trust, customer impact, technical debt, health |
| Craft form | personal essay, reported analysis, case study, visual essay, code tutorial |
Example expansions for `AI agents`:
- `AI agents production failure`
- `AI agent evaluation tutorial`
- `what I learned building AI agents`
- `AI agents human review`
- `AI agents startup operations`
- `multi-agent coordination`
- `AI agents trust and governance`
- `AI agents changed my role`
The final story should select one central tension, not combine every query.
## Story Expansion Lenses
### Lived or observed experience
What did the author see, choose, build, lose, test, misunderstand, or change? Which detail could not come from a generic search result?
### Reader transformation
What should the reader understand, feel, question, or do differently after the story?
### Fresh synthesis
Can two domains, reports, or experiences be connected in a way that reveals something useful and original?
### Failure and boundary
Where does the popular method fail? What is costly, emotionally difficult, risky, inaccessible, or context-dependent?
### Craft and form
Would the idea work best as a scene-led essay, tutorial, teardown, reported analysis, visual essay, annotated build, or comparison?
### Medium community fit
Which Topic and Publication communities already care about this question? What have they read repeatedly, and what is missing?
### Follow-up potential
Does the story open one coherent next question without withholding the current answer?
## 2026 Medium Topic Seeds
This map was refreshed on **2026-08-12** from Medium Help, Topic pages, recent recommended stories, and active Publication patterns. It is a discovery seed, not a popularity ranking. Recheck Topics and examples before every time-sensitive assignment.
### 1. Applied AI beyond the hype cycle
Core Topics and phrases:
- Artificial Intelligence;
- AI Agents and agentic workflows;
- context engineering and tool use;
- RAG, evaluation, and production reliability;
- multi-agent coordination;
- human review, governance, privacy, and trust;
- AI's effect on work and professional identity.
Story opportunities:
- A build log that shows where an agent failed and how the workflow changed.
- Why a strong demo became a weak production system.
- What human review actually costs.
- Moving from prompt tricks to context, tools, and evaluation.
- The emotional or organizational cost of automating judgment.
- A documented experiment comparing one agent pattern across real tasks.
Weak version to avoid: generic “top AI trends/tools” content with affiliate links and no original testing.
### 2. Data science, analytics, and decision quality
Core Topics and phrases:
- Data Science, Machine Learning, Statistics, Python;
- causal inference and experimentation;
- data visualization and communication;
- production ML, MLOps, and evaluation;
- data careers in an AI-shaped market;
- domain-specific analytics;
- misleading metrics and repeated statistical myths.
Story opportunities:
- The statistic everyone repeats but no one rechecked.
- A causal question that prediction cannot answer.
- A chart teardown tied to a real decision.
- What data science roles now require beyond modeling.
- A reproducible method with code, assumptions, and failure cases.
- How an analysis changed after domain experts challenged it.
Weak version to avoid: copied notebook steps without interpretation, provenance, or a reader decision.
### 3. Software engineering in the agentic era
Core Topics and phrases:
- Programming, Software Engineering, Web Development;
- AI-assisted coding and code agents;
- system design and architecture;
- reliability, observability, testing, and security;
- documentation and developer experience;
- legacy systems and durable “boring” technology;
- developer careers and the changing role of the IC.
Story opportunities:
- Where code generation shifts work rather than removes it.
- The test or observability signal that caught an AI-generated defect.
- Why specifications outperform intuition for one class of project.
- A legacy technology that remains valuable and why.
- How developers actually read and use documentation.
- A migration postmortem with transferable limits.
Weak version to avoid: a framework tutorial that invents production experience or omits version and environment context.
### 4. Product, UX, and the new digital divide
Core Topics and phrases:
- UX Design, Product Design, UI Design, UX Research;
- cognitive load and perceived performance;
- AI access, affordability, and inclusion;
- AI-assisted research and design workflows;
- accessibility and responsible defaults;
- design systems and interface sameness;
- case studies with real constraints.
Story opportunities:
- Why a visually simple interface still exhausts users.
- What AI affordability changes about product access.
- Where AI helps ideation and where it weakens it.
- A before/after decision backed by research rather than visual preference.
- A case study that includes what failed, not only the polished screens.
- How perceived speed changes user trust.
Weak version to avoid: a fictional UX case study presented as client work.
### 5. Careers, leadership, and changing work
Core Topics and phrases:
- Career Advice, Leadership, Management, Future of Work;
- skill shifts and portfolio evidence;
- individual contributor vs management paths;
- hiring uncertainty and role redesign;
- communication, mentorship, and cross-functional work;
- learning without endless course consumption;
- burnout, ambition, and professional identity.
Story opportunities:
- The career advice that stopped working after a role changed.
- A concrete artifact that proved capability better than a credential.
- What managers must do when output becomes cheaper to generate.
- The invisible ceiling or trade-off in a senior IC path.
- A leadership mistake told with honest consequences.
- How learning changed when it became tied to a real project.
Weak version to avoid: universal career rules based on one unexamined anecdote.
### 6. Entrepreneurship, product building, and operating risk
Core Topics and phrases:
- Entrepreneurship, Startup, Business Strategy;
- small or overlooked startup ideas;
- founder failure and decision-making;
- AI-native products, no-code, and vibe coding;
- customer discovery, pricing, and distribution;
- build vs buy and vendor dependence;
- operational risks hidden behind product demos.
Story opportunities:
- The small idea that became viable because a constraint changed.
- A founder decision that looked rational but failed.
- What a no-code or AI-built prototype could not solve.
- The person, vendor, or system that could stop the business with one failure.
- A pricing or distribution experiment with real limits.
- Why software that “half fits” costs more than its invoice.
Weak version to avoid: motivational founder mythology or unverified revenue claims.
### 7. Self-improvement, psychology, and relationships
Core Topics and phrases:
- Self Improvement, Psychology, Mental Health, Personal Growth;
- habits and productivity skepticism;
- identity, recovery, grief, aging, and life transitions;
- relationships, friendship, intimacy, and boundaries;
- mindfulness and attention;
- work-life tension and meaning.
Story opportunities:
- A specific personal experiment and what it did not fix.
- The cost of returning to a former identity or habit.
- Advice that failed during depression, grief, burnout, or transition.
- A relationship pattern understood through one honest scene.
- Why retirement or healing is not a productivity problem.
- A research-informed essay that keeps lived experience and clinical claims separate.
Weak version to avoid: medical or mental-health certainty without authority, sources, or safety boundaries.
### 8. Writing, creativity, and human voice
Core Topics and phrases:
- Writing, Creativity, Storytelling, Editing;
- human voice in the age of AI;
- revision, craft, and publishing practice;
- creative constraint and attention;
- authenticity without performance;
- visual storytelling and personal essays.
Story opportunities:
- What an AI-assisted draft could not know about the author's story.
- A revision that changed the meaning, not just the style.
- Create like a child, edit like a scientist—tested on a real project.
- Why voice comes from selection and consequence, not quirky wording.
- A transparent creative process with drafts and decisions.
- The line between assistance, derivation, and authorship.
Weak version to avoid: advice for “beating AI detectors” or imitating a living writer's distinctive voice.
### 9. Science, climate, society, and technology ethics
Core Topics and phrases:
- Science, Climate Change, Technology, Ethics;
- scientific misconceptions and verification;
- digital inequality and access;
- innovation vs social consequence;
- infrastructure, energy, and sustainability;
- policy translated into lived impact.
Story opportunities:
- A familiar statistic rechecked against the original evidence.
- How a technical system changes who can participate.
- A local observation connected carefully to a larger dataset.
- What one policy changes for a specific community.
- A scientific explanation built around a concrete question.
- A technology trade-off without utopian or dystopian certainty.
Weak version to avoid: overstating causality, certainty, or expertise in high-impact domains.
## Topic Selection Pattern
Medium allows up to five Topics. Use this composition only when each Topic is verified and supported:
| Role | Example for an AI UX story |
|---|---|
| Broad interest | Technology |
| Field/craft | UX Design |
| Specific subject | Artificial Intelligence |
| Reader intent/outcome | Product Design |
| Adjacent perspective | Accessibility |
Prefer three strong Topics to five weak ones.
Avoid:
- spelling variants that lead to tiny accidental Topic pages;
- Topics chosen only for follower count;
- current-year micro-Topics when the established Topic is more accurate;
- an unrelated broad Topic such as `Technology` on a story with no technology content;
- duplicative Topics that add no reader distinction.
## Publication Research
For each candidate Publication, record:
| Field | What to verify |
|---|---|
| Theme | What the Publication promises readers |
| Recent fit | Which recent stories overlap and what gap remains |
| Submission state | Open to all, following, contributing, closed, or unknown |
| Requirements | Image, subtitle, topic match, minimum read time, paywall, draft/published |
| Rights/canonical | Prior publication, exclusivity, editing, and canonical expectations |
| Voice and craft | Typical level of experience, evidence, code, narrative, or visuals |
| Status | Research only, eligible to submit, submitted, edits requested, approved, declined |
Do not submit automatically. Publication editors may edit stories and control acceptance; local completion is not acceptance.
## Candidate Story Scorecard
Score 1-5:
| Criterion | Question |
|---|---|
| Reader pull | Is there a clear curiosity, problem, emotion, or decision? |
| Author advantage | Does this author have real material or a defensible original synthesis? |
| Freshness | Is there a current change, overlooked detail, or new interpretation? |
| Evidence/material | Are scenes, examples, data, code, or sources available? |
| Medium fit | Does the idea match a real Topic or Publication audience? |
| Story movement | Can the idea progress rather than repeat one point? |
| Reader value | Will the reader's time be rewarded? |
| Policy fit | Can it comply with authorship, disclosure, rights, and distribution rules? |
Reject or convert to interview-first mode when `Author advantage` or `Policy fit` is below 3.
## `medium-topic-map.md` Template
```markdown
# Medium Topic and Story Map
## Research context
- Research date:
- Author:
- Target reader:
- Language:
- Existing audience/material:
- Medium access available:
- Verification limits:
## Query and observation log
| Query | Medium surface | Observed signal | External evidence | Author advantage | Interpretation |
|---|---|---|---|---|---|
| ... | Topic / Search / Publication / Staff Picks | ... | ... | ... | ... |
## Repeated questions and content gaps
1. ...
2. ...
3. ...
## Candidate stories
| Story | Reader tension | Author contribution | Topics | Publication fit | Score | Risk |
|---|---|---|---|---|---:|---|
| ... | ... | ... | ... | ... | 0-40 | ... |
## Selected story
- Working idea:
- Reader:
- Core insight:
- Author reason:
- Why now:
- Narrative/evidence material:
- Primary Topic:
- Supporting Topics:
- Publication recommendation:
- Explicitly out of scope:
- Author material still needed:
- AI/disclosure status:
## Refresh note
- What must be rechecked before submission or publication:
```
## Research Stop Rule
Stop when:
- the reader tension and story promise are specific;
- the author has real material or an original evidence-based angle;
- 3-6 material claims have viable sources;
- the Topic set is relevant and verified;
- one Publication or profile-first path is justified;
- the story has movement, a limit, and an ending direction;
- policy and disclosure needs are clear.
More trend browsing after this point usually encourages imitation rather than originality.
## Official Medium Reference Points
Recheck these pages when platform behavior matters:
- Distribution and Boost/General/Network guidance: https://help.medium.com/hc/en-us/articles/360006362473-Medium-s-Distribution-Guidelines-How-curators-review-stories-for-Boost-General-and-Network-Distribution
- AI content and disclosure policy: https://help.medium.com/hc/en-us/articles/22576852947223-Artificial-Intelligence-AI-content-policy
- Using up to five Topics: https://help.medium.com/hc/en-us/articles/214741038-Using-topics
- Search and homepage discovery: https://help.medium.com/hc/en-us/articles/115012586467-Your-homepage
- Publication submission and requirements: https://help.medium.com/hc/en-us/articles/213904978-How-to-submit-a-story-to-a-publication
- Publishing and preview settings: https://help.medium.com/hc/en-us/articles/225168768-Writing-and-publishing-your-first-story
- Detailed story Stats and reader interests: https://help.medium.com/hc/en-us/articles/34831991136151-Story-s-detailed-stats-page
- Canonical links: https://help.medium.com/hc/en-us/articles/360033930293-Set-a-canonical-link
- Importing a previously published story: https://help.medium.com/hc/en-us/articles/214550207-Importing-a-post-to-Medium
These sources describe current platform behavior. They do not guarantee distribution, indexing, Publication acceptance, reads, or earnings.
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| rea-analysis-optimizationmorluto/reaOptimize REA providers and shared analysis workflows for agent context, latency, CPU, and memory while preserving correctness and Evidence. Use for analysis bottlenecks and cache or query design; CI/test-suite optimization belongs to rea-ci-tests-audit. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-ci-tests-auditmorluto/reaAudit REA CI and test suites for correctness, fixture drift, false passes, and measured performance; use for suite or workflow investigations and efficiency changes, rather than ordinary feature testing or a single test failure. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| rea-tool-designmorluto/reaAudit, investigate, design, or change REA investigation tools, CLI/MCP contracts, provider capabilities, and Evidence semantics. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| ponytail-auditDietrichGebert/ponytailQuality audit of the whole repo: bugs, security, real load, missing tests, speed, and what to delete. Most important first. | Stars: 194k | Updated | Claude CodeCodex | Verified | Install |