comparison
Skill1 starsChanged 6 months agoApache-2.0View on GitHub
Generate a head-to-head comparison page ("[Site] vs [Competitor]" or "[Product A] vs [Product B]") for a chosen managed site. Captures decision-stage searchers comparing the site against a named alternative. Picks the comparison from `service-keywords.csv` (rows with intent=`comparison`) or user input, runs SERP analysis on top 3 comparison pages, builds a feature-by-feature comparison table, balances honesty (real limitations) with conversion (clear differentiators), satisfies Tier 1 on-page SEO, and ships per the site's publishing method. Use when the user types `/comparison` or asks for a "vs" page.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- aspose-words/agentic-net-examplestheir own version, first seen Sep 9, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- aspose-words/agentic-net-examples★ 4same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository20 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- alternativeSkill
- auditSkill
- blogSkill
- broken-backlinksSkill
- case-studySkill
- clusterSkill
- competitor-backlinksSkill
- glossarySkill
- haroSkill
- integrationsSkill
- lovable-deploySkill
- pricingSkill
- programmatic-batchSkill
- refreshSkill
- semrush-baselineSkill
- serp-featuresSkill
- serviceSkill
- triageSkill
- wireframeSkill
The file
We split this skill into the rules it gives an agent
4 rules. About 0 are common, 1 is distinctive, and 3 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
3
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 39
Every claim about the competitor in the final page MUST cite one of the four sources above.
- Very specific: line 55
Do NOT use Product schema for the competitor's product — only mentions / about.
- Very specific: line 58
Every claim about the competitor MUST be sourced (linked, even if nofollow noopener).
- Very specific: line 59
Every claim about our offer MUST be backed by stats.md or a story from stories.md.
≈ “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.
Files
/comparison — Head-to-head comparison page generator
Creates a decision-stage comparison page. Comparison search intent is the highest-conversion intent on the web — searchers are pre-qualified, already evaluating named options, and ready to choose. This skill captures that traffic without sounding like a hatchet job (one-sided comparisons rank poorly and convert worse).
Inputs
Required (asked at start if not given):
- Site — must match a folder under
sites/[name]/. - Comparison target — either a primary keyword (
site-a vs accenture ai consulting) or{competitor}slug. The skill picks fromservice-keywords.csvrows with intent=comparisonif not supplied.
Optional (per-run overrides):
- Competitor data source — defaults to web research; override to a supplied dossier file in
_research/competitors/{slug}.md. - Stance —
balanced(default — honest pros/cons both sides) |differentiator-led(lead with our edge, still honest about limitations). Neverhatchet— refused.
Workflow
This skill follows /service's workflow (Steps 0–11) with the page-type-specific differences listed below. Read .claude/skills/service/SKILL.md end-to-end before running.
What's unique to /comparison
Step 2 — Keyword selection: pick from service-keywords.csv rows where intent=comparison, or accept user-supplied keyword. Validate via SERP comparison test (SEO_GUIDE.md Section 2.4) — comparison pages must compare exactly the two named entities, not a broader category.
Step 4 — SERP analysis (comparison-specific):
- Top 3 comparison pages — extract the comparison axes (price, feature set, support, integrations, learning curve, etc.). The shared axes across all 3 define the table this page MUST cover.
- Identify which axes the top 3 OMIT — those are differentiation opportunities for this page.
- Note tone: is the top result a balanced third-party review (Capterra, G2, TechRadar) or a competitor-owned page? Third-party-dominated SERP = trust signal is hard; pace conversion CTAs accordingly.
Step 5 — Competitor research: for the competitor side of the comparison, gather (in this order):
- Their own public pricing page + features page (URL + date accessed)
- G2 / Capterra / TrustRadius reviews (last 90 days) — pull 3 verbatim pain points
- Their public case studies — what they explicitly target (industry, company size)
- Independent third-party reviews — confirm or refute their own claims
Every claim about the competitor in the final page MUST cite one of the four sources above. No paraphrased hearsay. No "competitors often struggle with X" without a citation.
Step 6 — Wireframe (comparison-specific):
- Hero zone: "[Our offer] vs [Competitor]: which fits [decision criterion]?" — frames the page as decision support, not a sales pitch.
- Quick answer (above the fold): 2-sentence verdict — "If you need X, pick us. If you need Y, pick them." Honest. Pre-qualifies the reader.
- Comparison table (the spine): 6–12 rows of feature-by-feature comparison. Each cell: factual (✓/✗ or short fact). No marketing copy in cells.
- Section per major axis: 200–400 words each. Explain the trade-off, name the use case for each side. End with "When [our offer] wins" and "When [competitor] wins" subheadings.
- "When NOT to choose us" section (required, not optional). 2–3 specific scenarios. This is the trust-building section that makes the rest of the page believable.
- FAQ: 5–8 Qs covering: "Is this comparison biased?", "Can I switch later?", "What about [third option]?", pricing-specific Qs.
- Final CTA: softer than service pages — "See if [our offer] fits your team" rather than "Get a demo today." Comparison-stage readers reject hard sells.
Step 7 — Schema (comparison-specific):
Articlewithabout: [ {@type: Product, name: "Our Product"}, {@type: Product, name: "Competitor"} ]FAQPageBreadcrumbList- Optionally:
RevieworItemListreferencing both entities (used only if the comparison spans multiple options, not 1v1). - Do NOT use
Productschema for the competitor's product — onlymentions/about. Schema-asserting facts about another company's product is risky.
Step 8 — Voice + AI-search (comparison-specific extras):
- Every claim about the competitor MUST be sourced (linked, even if
nofollow noopener). - Every claim about our offer MUST be backed by
stats.mdor a story fromstories.md. - One opinion max — usually "[Differentiator] is what we'd pick on for [use case]" backed by a stat.
- AI-search rules: Q+A density (5+ in-body Q→A — comparison content is high-LLM-citation material).
Refusal conditions
This skill refuses to ship if:
- The comparison would require fabricating facts about the competitor (no public source available).
- The site's
voice.mddoesn't permit a comparison page (some brands forbid named competitor mentions). - A
/comparisonpage for the same competitor already exists inused-keywords.md(refresh instead via/refresh). - Stance is
hatchet— comparison must be balanced or differentiator-led, never one-sided.
Output
Standard _drafts/[slug]/ layout (or /[lang]/ subfolders for multilingual sites), with these comparison-specific additions:
comparison-table.md— the raw table as a separate file (some publishing flows need it as a structured asset)competitor-sources.md— every citation used, with URL + date accessed (the receipts file)
Length target: 1,800–3,000 words (SERP-driven; comparison pages skew longer than blog posts because the table + per-axis breakdowns are mandatory).
Routine versioning + idempotency contract
Same as /blog (see .claude/skills/blog/SKILL.md). Output path is _drafts/{slug}/ per language. Idempotency key: (site, slug, language). Skip re-run if _drafts/{slug}/ exists with a meta.json containing today's date.
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
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.

