Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add semperi/restaurant-marketing-skills --skill analyze-restaurant-reviewsgit clone --depth 1 https://github.com/semperi/restaurant-marketing-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/semperi/restaurant-marketing-skills/analyze-restaurant-reviews)<a href="https://agentmods.dev/skills/semperi/restaurant-marketing-skills/analyze-restaurant-reviews"><img src="https://agentmods.dev/badge/skills/semperi/restaurant-marketing-skills/analyze-restaurant-reviews/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/semperi/restaurant-marketing-skills/analyze-restaurant-reviews"><img src="https://agentmods.dev/badge/skills/semperi/restaurant-marketing-skills/analyze-restaurant-reviews.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00077 | $0.01821 |
| Opus 5 | $0.00039 | $0.00911 |
| Sonnet 5 | $0.00015 | $0.00364 |
| Haiku 4.5 | $0.00008 | $0.00182 |
Grade A, and why
analyze-restaurant-reviews scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 152 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Analyze Restaurant Reviews
You get the three to five things guests actually keep saying, each one carrying the guests' own words so you can see it for yourself. Not a sentiment score. Not a percentage. The themes, the quotes behind them, and the one worth looking at first.
What to paste
| What | Where to find it | If you don't have it |
|---|---|---|
| Twenty or more reviews | Google Business Profile → Reviews, or an export from your review tool | Fewer is fine — say how many you read and that themes from a small pile are thin |
| The star rating on each | Beside each review | Work from the words alone and say so |
| The date on each | Beside each review | Report themes only and skip the trend section entirely |
| Reviews from other platforms | Yelp, Tripadvisor, a delivery app | Google alone is fine — name which platform you read |
| What the owner already suspects | The owner | Read without it; a blind read is often more useful |
Everything in that table is the whole truth available to you — see Never do this below.
Quick start
Owner: "here's 40 reviews from the last six months, what's the pattern"
→ Read all 40. Group what repeats. Discard what appears once.
→ Name 3-5 themes. Quote 3-8 of the guests' own words under each one.
→ If dates are present, say what is growing and what is fading.
→ Point at the one thing worth looking at first. Stop there — do not fix it.
Workflow
Step 1 — Read everything before deciding anything
Read the whole pile first. The temptation is to name a theme after the fourth review and then spend the rest confirming it. Count how many reviews you actually read and say the number, because a read of eight reviews and a read of eighty support very different claims.
Step 2 — Group what repeats, discard what does not
A theme is something several guests independently raised. One person's strong opinion is not a theme, however memorable it is.
- Three to five themes. Do not pad to five. If the reviews only support two, name two and say the pile is thinner than it looks.
- Split praise from complaint. "The staff" is not a theme when half call them warm and half call them slow. Those are two themes.
- Name it in guests' language, not in business language. "Waits at peak" beats "service efficiency concerns".
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 152 lines · 77 tokens per session scan A 1b36bd0bc894
analyze-restaurant-reviews is a skill published in the GitHub repository semperi/restaurant-marketing-skills (1 stars, last pushed 1mo ago), licensed MIT. It adds 77 tokens to every session and 1,821 once invoked, about $0.0004 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
restaurant-local-seo
Local SEO audit — "best [cuisine] near me" targeting, Google Business Profile optimization, NAP consistency, citation building, restaurant schema markup.
exploratory-data-analysis
Perform bounded, local exploratory analysis of explicitly supported scientific files. Use for redacted CSV/TSV/JSON profiles; optional NumPy, HDF5, FASTA/FASTQ, and basic image metadata inspection; missingness/leakage audits; outlier and transformation sensitivity; and rigorous EDA report scaffolds. Other domain…
google-ads-audit
Google Ads account audit and business context setup. Use for account-health audits and business-context setup. Trigger on "audit my ads", "ads audit", "set up my ads", "onboard", "account overview", "how's my account", "ads health check", "what should I fix in my ads", or when the user is new to NotFair and hasn't run…
data-charts-tako
Search and visualize the world's data - get charts, insights, and embeddable knowledge cards for finance, economics, demographics, sports, and more.
webhook-management
Configure and validate CCAM webhook targets across supported chat, incident, automation, and generic providers. Use when listing provider requirements, creating or updating a target, scoping it to alert rules, sending a test notification, reviewing delivery history, or deleting a target.
gesellschaftsrechtliche-satzungen-agb
Für Gesellschaftsrechtliche Satzungen AGB Abgrenzung: ordnet Norm, Beweislast und Gegenargument; Ergebnis: Prüfprodukt mit Risiko und nächstem Schritt. Fachgebiet: AGB-Recht-Prüfer. Route: gesellschaftsrechtliche-satzungen-agb.