claude-desktop/ai-discoverability/SKILL.md
Submit your site to llms.txt directories and build backlinks from AI-indexed sources. Use to improve visibility in AI search results.
npx skillsauth add kubical-ai/aeo-geo-optimization ai-discoverabilityInstall this skill globally with one command. Works with Claude Code, Cursor, and Windsurf.
3 of 9 scanners reported clean
Some scanners were skipped, did not run, or reported a non-clean status. Review each row below.
Helps documentation sites get discovered and cited by AI systems through directory registration and external authority building.
Load
references/directories.mdfor the current list of llms.txt directories and submission details.
Before starting registration, verify:
https://[site-url]/llms.txt — open it in a browser to confirmIf llms.txt is missing, stop and note: Item #1 must be completed before #25.
Read references/directories.md for the current directory list. For each directory:
llmstxthub.com
llmstxtdirectory.org
thedaviddias/llms-txt-hub (GitHub)
Produce a checklist the user can work through:
## llms.txt Directory Submission Checklist
Your llms.txt is at: https://[site-url]/llms.txt ✓
### Directories to Submit To
- [ ] **llmstxthub.com** — [direct link to submission form]
- Required: Site URL, description, category
- [ ] **llmstxtdirectory.org** — [direct link to submission form]
- Required: Site URL, llms.txt URL, one-sentence description
- [ ] **thedaviddias/llms-txt-hub** — Submit a PR
- Fork: https://github.com/thedaviddias/llms-txt-hub
- Add to README: `| [Site Name] | [Site URL] | [llms.txt URL] | [Category] |`
Estimated time: 15-30 minutes
Use WebSearch to understand the current external presence:
Search: "[product name] review" site:g2.com OR site:capterra.com
Search: "[product name] github"
Search: "[product category] comparison blog" 2024 OR 2025
Search: "[product name] integration" OR "[product name] tutorial"
Note what exists and what gaps there are.
Based on the project (e.g., WMS/OMS/3PL for ShipEdge), identify:
Use WebSearch to find:
Create a prioritized action plan:
## External Authority Building Plan for [Project]
### Tier 1 — Quick Wins (1-2 weeks)
**Software Directories (High AI Training Data Coverage)**
1. **G2 Listing** — g2.com/products/[product]/reviews
- Impact: High — G2 pages are heavily indexed by AI training data
- Action: Create/complete listing with detailed description, features, use cases
- Est. time: 2-3 hours
2. **Capterra Listing** — capterra.com
- Impact: High
- Action: Submit product listing
- Est. time: 1-2 hours
3. **[Niche-specific directory]** — [URL]
- Action: [specific steps]
### Tier 2 — Content & Community (1-2 months)
**Open Source Integration**
- Create a minimal example repo: `github.com/[org]/[product]-examples`
- Shows real API usage
- Gets indexed by GitHub search + AI training data
- Action: Build a simple "Hello World" integration in the most popular platform (e.g., Shopify)
**Developer Blog Post**
- Target publications: [research and list 2-3 relevant blogs]
- Topic idea: "[How we solved X] with [product]" — a technical case study
- Impact: Developer blogs are heavily weighted in AI training data
**Stack Overflow Presence**
- Tag to watch: [relevant tag]
- Action: Answer 5-10 questions where the product is relevant
- Adds authoritative mentions in a heavily-indexed source
### Tier 3 — Long Term (3-6 months)
**Community Building**
- [Discord/Slack community if not existing]
- [Newsletter targeting developers in the space]
**Partnership Content**
- Guest posts on [integration partner] blogs
- Co-marketing with [complementary tools]
### Metrics to Track
- G2 rating and review count
- GitHub stars on example repos
- Mentions in "best [category]" lists when asked to AI tools
- Direct: ask ChatGPT/Perplexity "What is [product]?" and note what it says
Recommend a simple way to track AI mentions:
Present results clearly split into two sections:
Note: most of #26 requires the user to take action (write content, create accounts, etc.) — your role is to research and plan, not execute the outreach directly.
documentation
Create llms.txt and llms-full.txt files following the llmstxt.org spec. Use when setting up AI discoverability files for a docs site.
tools
Create AI editor context files (.cursorrules, CLAUDE.md, Copilot instructions) for a project. Use when setting up AI coding tool integration.
testing
Audit and optimize a docs site for AI citation. Scans 29 AEO/GEO signals, plans fixes, and implements them with approval at each step.
testing
Create or fix llms.txt and llms-full.txt files for a documentation project, following the llmstxt.org specification. Use this skill whenever someone needs to create an llms.txt file, generate a full documentation text dump (llms-full.txt), or fix a malformed llms.txt. Also use when optimizing a docs site for AI crawlers, when setting up AI discoverability files, or when the user mentions making their docs readable by LLMs, ChatGPT, Perplexity, or other AI systems.