ObsurfableShort answer: Developer-tools buyer prompts do not produce one shared product shortlist across AI engines. ChatGPT most often opens with headless CMS and platform catalog tools — Contentful is the #1 pick on 6.1% of category prompts and Backstage on 5.9%, and 17.1% of ChatGPT #1 picks fall in a CMS/platform bucket. Grok opens with IDE and delivery-stack tools — 41.5% of its #1 picks are GitHub, VS Code, or CI products, led by GitHub at 8.9%. Claude and Gemini skew toward form builders, with SurveyJS as Claude’s #1 on 14.7% of prompts and mentioned on 38.8% of Claude answers versus 1.5% on Grok. On 134 matched ChatGPT–Grok prompts, the two engines name a different #1 brand on 70.9% of questions.
The disagreement is not a tie-breaker between two CRMs. It is different product classes inside the same category label.
What we analysed
We used Obsurfable Explorer’s research corpus: buyer-style prompts tagged Technology · Developer Tools — forms, headless CMS, developer portals, PDF/API utilities, CI, and adjacent builder tooling.
For each prompt we keep the latest observation per platform family (ChatGPT, Gemini, Claude, Perplexity, Grok), preferring web-app captures over API runs when both exist. We report:
- #1 pick — first brand extracted from the answer
- Mention rate — brand appears anywhere in the extracted list
We grouped a subset of #1 picks into CMS/platform catalog (Contentful, Sanity, Backstage, WordPress, Strapi, and peers) and dev workflow (GitHub, VS Code, GitHub Actions, CircleCI, Cursor, and peers) to show when engines answer different types of tools on the same category tag.
| Scope | Value |
|---|---|
| Corpus window | 14 July – 28 October 2026 |
| Developer Tools prompts | 1,537 |
| Observations in category | 2,531 |
| ChatGPT + Grok matched panel | 134 |
| ChatGPT + Claude matched panel | 231 |
| ChatGPT + Gemini matched panel | 228 |
Browse live views on Developer Tools, ChatGPT in Developer Tools, Grok in Developer Tools, and Claude in Developer Tools. See corpus profiles for Contentful, GitHub, and SurveyJS.
What we found
1. ChatGPT favors CMS and catalog tools; Grok never leads with them
Category-wide #1 picks (latest observation per prompt per engine):
| Engine | Prompts observed | Top #1 brand | Share as #1 | CMS/platform #1 share | Workflow #1 share |
|---|---|---|---|---|---|
| ChatGPT | 1,536 | Contentful | 6.1% | 17.1% | 6.9% |
| ChatGPT | 1,536 | Backstage | 5.9% | (included above) | — |
| Grok | 135 | GitHub | 8.9% | 0.0% | 41.5% |
| Grok | 135 | Visual Studio Code | 5.9% | — | (included above) |
| Claude | 232 | SurveyJS | 14.7% | 0.4% | 0.4% |
| Gemini | 228 | SurveyJS / React Hook Form | 7.5% each | 0.0% | 0.4% |
Mention rates show the same split with larger gaps on CMS brands:
| Brand | ChatGPT mention | Grok mention | Claude mention | Gemini mention |
|---|---|---|---|---|
| Contentful | 11.1% (170/1,536) | 0.0% (0/135) | 0.0% | 0.0% |
| Sanity | 10.6% (163/1,536) | 0.0% | 0.0% | 0.0% |
| Backstage | 8.7% (133/1,536) | 2.2% (3/135) | 0.9% | 1.3% |
| GitHub | 10.8% (166/1,536) | 20.7% (28/135) | 1.7% | 2.2% |
| SurveyJS | 2.9% | 1.5% | 38.8% (90/232) | 32.5% (74/228) |
Grok’s developer-tools answers never mention Contentful or Sanity in this window, while ChatGPT mentions both on more than one in ten prompts. That is a structural visibility gap, not a rank swap between two interchangeable vendors.
2. Claude and Gemini share a form-builder house list ChatGPT and Grok barely use
Claude and Gemini concentrate #1 picks on SurveyJS, React Hook Form, Typeform, and adjacent form products:
| Engine | SurveyJS #1 | React Hook Form #1 | Typeform #1 |
|---|---|---|---|
| Claude (n=232) | 14.7% | 6.5% | 4.7% |
| Gemini (n=228) | 7.5% | 7.5% | 6.6% |
| ChatGPT (n=1,536) | 1.3% | 1.3% | 1.6% |
| Grok (n=135) | 0.7% | 0.7% | 0.0% |
On matched panels where the same prompt was observed on two engines:
| Panel | n | Different #1 brand |
|---|---|---|
| ChatGPT + Grok | 134 | 70.9% |
| ChatGPT + Claude | 231 | 61.0% |
| ChatGPT + Gemini | 228 | 56.6% |
So even when the question text is identical, the #1 product often changes — and the direction of change depends on which engine pair you compare.
3. Grok stacks IDE/CI #1 picks; ChatGPT spreads across CMS, forms, and PDF utilities
Among Grok answers with an extracted #1 brand, 41.5% lead with workflow tools (GitHub, Visual Studio Code, GitHub Actions, CircleCI, Cursor, Harness, and peers). The next-largest Grok #1 clusters are GitHub Actions (5.2%) and VS Code variants combined (~10.4%) alongside GitHub (8.9%).
ChatGPT’s top #1 list is flatter but CMS-heavy at the top: Contentful (6.1%), Backstage (5.9%), then GitHub (4.2%), Sanity (1.8%), WordPress (1.6%), Typeform (1.6%), and PDF/API utilities (Adobe PDF Services, Puppeteer, Apryse) each around 1.5–2.8%. 21.9% of ChatGPT observations yield no extracted #1 brand — higher than Grok (0.7%).
Why this is surprising
“Developer tools” reads like one market map. In the corpus it behaves like several overlapping markets that different engines default to:
- ChatGPT surfaces CMS and internal-developer-portal vendors (Contentful, Backstage, Sanity) on buyer prompts that share a category tag with form and CI questions.
- Grok defaults to the delivery toolchain (GitHub, editors, CI) and does not surface major CMS names at all.
- Claude and Gemini default to form libraries (SurveyJS in particular) on the same label — a pattern closer to “form builder recommendations” than to Grok’s GitHub-first list.
A visibility program that tracks one engine, or one keyword cluster inside Developer Tools, can show strong Share-of-voice for a brand that another engine never places in-category.
Limitations
- Category breadth: Developer Tools mixes forms, CMS, PDF tooling, CI, and API utilities. We report engine defaults on the full tagged set, not a single sub-topic.
- Coverage imbalance: ChatGPT has 1,536 latest observations in this window versus 135 for Grok and 228–232 for Gemini and Claude. Rankings for smaller panels are noisier; mention-rate gaps on CMS brands are still extreme (0% vs >10%).
- Brand extraction: #1 and mention rates depend on structured brand extraction from free-text answers; 21.9% of ChatGPT rows have no #1 brand.
- Perplexity: Only six Perplexity observations met the latest-per-prompt rule in this category during the window — too small to compare fairly; we omit Perplexity from engine tables above.
Methodology
- Source: Obsurfable Explorer research corpus (buyer-style prompts in the public research set).
- Platforms: ChatGPT, Gemini, Claude, Perplexity, Grok — latest observation per prompt per family, web UI preferred over API when both exist.
- Window: Observations recorded 14 July – 28 October 2026.
- Metrics: #1 brand = first extracted mention; mention rate = any extracted mention on the latest answer; CMS/workflow buckets = predefined brand sets listed above.
- Matched panels: Prompts with latest observations on both listed engines within the category.
Conclusion
Developer-tools AI visibility is engine-specific and product-class-specific. ChatGPT elevates Contentful, Backstage, and Sanity; Grok elevates GitHub, VS Code, and CI while never mentioning Contentful or Sanity; Claude and Gemini elevate SurveyJS and form stacks on 39% and 33% of their answers respectively. Treating “Developer Tools” as one leaderboard — or optimizing for a single engine — will miss where each engine’s house favorites actually live.
Explore the data: Developer Tools · ChatGPT · Grok · Claude