skill
local-service-research
Claude skill: turns Facebook group word-of-mouth into a sourced, ranked comparison of local businesses
Set up
Skills are installed by placing their folder where your agent loads skills.
The download comes from github.com/ann-703/local-service-research itself, not from our server, so what you get is exactly what its author published.
Install instructions come from the README
README
synced Oct 23, 2026Local Service Research — a Claude Skill
Turn a name (or a list of names, or "find me a good X near Y") into a clear, sourced recommendation about which local business to use — for any category: massage & spas, plumbers, electricians, contractors, vets, daycares, salons, cleaners, dentists, mechanics, movers, and more.
The skill gathers reviews, ratings, and pricing from the web, writes a structured note for each place, and rolls everything into a single shareable comparison report.
What it produces
- The answer, inline — the verdict (who to pick and why) and a Markdown comparison table, right in the conversation. You don't open a file to find the recommendation.
- One research note per place (
.md) — details, ratings (with review counts, cross-sourced), pricing, pros, cons/red flags, and a plain-language verdict. Template:assets/research-note-template.md. - A Markdown comparison file (
.md) — readable in any editor, the chat, or on your phone. The default, reviewable artifact. - An optional HTML report for sharing — a styled, self-contained file you
can open and Print → Save as PDF to email along. Produced on request, not
by default. Template:
assets/report-template.html.
No external dependencies — it does not require a PDF engine, headless browser, or any paid API.
Install
This is a Claude Skill.
Claude Code (CLI / desktop):
git clone https://github.com/ann-703/local-service-research.git \
~/.claude/skills/local-service-research
Or copy the local-service-research/ folder into ~/.claude/skills/. Restart
your session and the skill auto-loads.
Claude.ai / Cowork: upload the packaged .skill file, or add the folder to
your skills directory per your environment's instructions.
Use
Just talk normally. The skill triggers when you're evaluating local providers — you don't have to name it. Examples:
- "Is Aroma Spa near Juanita Village any good?"
- "Compare these three plumbers and tell me who to call."
- "Find me a well-reviewed vet in Ballard with weekend hours."
- "Should I take my car to this shop? Here's the name."
It will ask where to save the research the first time (no paths are hardcoded), then build notes and a report you can share.
How it works
| File | Purpose |
|---|---|
SKILL.md | The skill: workflow, note structure, reporting style, guardrails |
assets/research-note-template.md | Per-place note template |
assets/report-template.html | Self-contained comparison report (print to PDF) |
references/sources-and-scoring.md | Where to look, how to weigh ratings, category-specific criteria |
Design principles
- Verdict first — people want the answer, then the reasoning.
- Cite sources — every claim is checkable.
- Honest about evidence — few reviews, no pricing, or one loud complaint is flagged, not glossed over.
- No fabrication — never invents ratings, prices, or quotes.
- Portable — asks for an output folder; no machine-specific paths.
License
MIT — see LICENSE. Use it, fork it, adapt it.
Credits
Built with Anthropic's skill-creator. Contributions welcome via PR.