restaurant picker
Rank restaurants based on your personal taste profile and flavor preferences instead of generic star ratings.
$5
Works with the AI Agents you already use
Claude Code
Cursor
Codex CLI
DeepSeek
Muse+22 moreClaude CodeCursorCodex CLIDeepSeekMuse
Dots
Manus
Grok
Grok Bot+18 agentsrestaurant picker
Example session with this skill installed
I'm in London and want to try a new pasta place. Check Padella, Bancone, and Trullo against my profile. I'm with one other person, casual dinner, budget is £50 total.
- Read your context and instructions
- Compiled the restaurant picker
I have analyzed Padella, Bancone, and Trullo for a casual dinner for two. I've assumed a walk-in or short-notice booking is preferred, though Padella usually requires a virtual queue.
Restaurant Ranking for London Pasta
| Rank | Name | Tier | Most Similar | Dishes to Order | Warning Signs | Buzz | Evidence |
|---|---|---|---|---|---|---|---|
| 1 | Bancone | Likely to Love | Pastaio | Silk Handkerchiefs, Ox cheek mafalde | Can be loud/cramped | High | High |
| 2 | Padella | Worth a Try | Trattoria Gloria | Pici cacio e pepe, Pappardelle | Long queue times | Very High | High |
| 3 | Trullo | Worth a Try | River Cafe | Beef shin ragu | Requires booking ahead | Medium | High |
Analysis
Reviewers consistently mention that Bancone's Silk Handkerchiefs have a distinct, rich walnut butter profile that aligns with your preference for nutty, umami-forward sauces. It is closer to your favorite, Pastaio, than to the bland chain options you've disliked.
Padella offers exceptional quality, but the high volume of "rushed service" reports in recent reviews matches a red flag in your profile. While the pici is a signature you would enjoy, the experience might feel too transactional for a relaxed casual dinner.
Recommended: Bancone (Covent Garden). It hits your preference for handmade texture and specific regional sauces without the extreme queue friction of Padella.
Next steps
- Would you like me to check real-time availability for Bancone?
- Shall I add Padella to your "Pending Feedback" list for when you visit?
- Do you want to see if any other "Likely to Love" spots are nearby?
Connects securely to your tools. The creator never sees your data.
About this skill
Crowd-sourced star ratings often fail because they do not account for your specific palate. This skill shifts the focus from global averages to personalized flavor profiles, ranking restaurant candidates by comparing their signature dishes against your own established benchmarks. It uses a systematic evidence-gathering process to identify "likely to love" spots while filtering out specific red flags defined in your taste history.
What it does
- Personalized ranking categorizes restaurants into four tiers based on your specific green and red flags.
- Flavour-centric analysis extracts descriptions of signature dishes from reviews rather than relying on star counts.
- Benchmark comparison matches new candidates against your favorite and least favorite places in the same category.
- Continuous learning updates your taste profile after every meal to refine future predictions.
- Buzz integration factors in the social value of famous or much-discussed places if specified in your profile.
How it works
- Load taste profile by retrieving your benchmarks, red flags, and want-to-try list from memory or local files.
- Clarify requirements for the specific occasion, including location, party size, and budget.
- Analyze evidence by scanning recent and negative reviews for specific flavor profiles and craft signals.
- Assign tiers through pairwise comparison with your benchmark restaurants.
- Generate report including a ranked table with dish recommendations and specific warning signs.
Frameworks & tools
The skill is designed for agents with web browsing capabilities to access recent review data from Google Maps, Yelp, or local platforms like Tabelog and Naver. It requires a markdown-based taste profile for persistent memory.
Why this beats prompting it yourself
Standard prompts usually result in generic "best of" lists that mirror biased public opinion. This skill enforces a strict "no-score" comparison method that forces the agent to justify rankings based on your documented flavor preferences and past experiences.
Use cases
- Deciding between three competing ramen shops based on noodle texture preferences.
- Screening a new fine-dining spot for specific red flags like "generic seasoning" or "inflated hype."
- Building a customized "want-to-try" list for an upcoming trip to a new city.
- Troubleshooting why a highly-rated restaurant failed to meet your expectations.
Known limitations
Does not predict exact numerical scores. Accuracy depends on the availability of descriptive reviews rather than just star ratings. Cannot access private saved lists or accounts.
Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for Agents & Orchestration.
Requires descriptive flavor reviews to work effectively. Cannot predict exact star ratings. Requires an initial setup of at least 15 rated places for high accuracy.
How to use
- 1
Connect your agent to the Agensi MCP
Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup
- 2
Sign in to Agensi
Your agent shows a sign-in link. Sign in and click Allow.
- 3
Ask for what you need
Your agent loads this skill and uses it on your task. It always asks before buying a skill.
Reviews
No reviews yet
Be one of the first to try it. Every listed skill goes through the trust checks below.
Security scanned
Checked by our 8-point scan before listing
Fresh listing
Recently published to Agensi
Money-back guarantee
Doesn't do what its page promises? You get your money back.
Trust & safety
Security scan
Security scan passed today