product-research
Find and compare current FF&E candidates from a design brief, with sourced purchasing facts and optional CSV-library save. Use to research or source products; not to match from an image or pair coordinating items.
Install / Use
npx skills add AlpacaLabsLLC/skills-for-architects --skill product-researchInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Data & AnalyticsSupported Platforms
Our assessment of product-research
product-research scores 89/100 on our quality scale, 343rd of 599 Data & Analytics skills we index.
Its SKILL.md is 15 KB long, well organised into 19 sections with 3 code examples: a thorough specification that gives an agent plenty to work with.
It has 361 GitHub stars, a meaningful sign that others use it.
Maintenance, license and trust
- The repository was last updated about 2 months ago, so product-research is actively maintained.
- Our last check on 2026-10-31 found the source still online.
- It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
Safety scan
No issues foundOur scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful.
AI review by kimi-k2.7-code on 2026-10-06. Automated pattern scan on 2026-10-05. It catches known dangerous patterns, not every risk — read a skill before letting an agent act on it.
product-research compared with similar skills
All 4 of these similar skills score higher than product-research; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| product-research (this skill)by AlpacaLabsLLC | 89 | 361 | 59d ago | SKILL.md |
| taste-skillby Leonxlnx | 100 | 118.9k | 11d ago | CLAUDE.md |
| Claude-Code-Game-Studiosby Donchitos | 100 | 40.2k | 10d ago | CLAUDE.md |
| LocalAIby mudler | 100 | 50.5k | today | MCP Server |
| algorithmic-artby anthropics | 100 | 177.9k | 58d ago | SKILL.md |
Frequently asked questions
- How do I install product-research?
- Run
npx skills add AlpacaLabsLLC/skills-for-architects --skill product-research. The install tabs above show the steps for each supported agent. - Which AI agents does product-research work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is product-research safe to use?
- Our scan of the whole file found no instruction hijacking, hidden characters, credential access, data exfiltration or destructive commands. An AI review of the same text found nothing harmful. It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is product-research still maintained?
- The repository was last updated about 2 months ago, so product-research is actively maintained.
Skill content
View source on GitHubname: product-research description: "Find and compare current FF&E candidates from a design brief, with sourced purchasing facts and optional CSV-library save. Use to research or source products; not to match from an image or pair coordinating items." allowed-tools:
- Read
- Write
- Edit
- Bash
- Glob
- Grep
- WebFetch
- WebSearch
- AskUserQuestion
/as:product-research — Product Research
Before acting, read the host contract and this component's declaration (skill:product-research). Load only its referenced mode profiles from the shared catalog. Compose modes required by the actual task; declarations are requirements, not proof of access or permission.
Library changes are owned by /as:product-library. Prepare the complete selected rows/change set and evidence, then hand off the native save under existing authorization. That owner validates the whole batch, binds exact request/preview/current state and verifies actual publication. This skill does not independently mutate product-library.csv.
Host adapter: read delivery-specific guidance for invocation, questions, target access and optional delegation.
Record authority and host handoff
For structured evidence handoff, read the complete native product-observation owner and schema. Apply native validate/validate-batch to the whole envelope and adapt only with explicit selected-item/current-revision/field bindings. Preserve exact typed values, source/locator/time and unknown status; retain full envelopes, audit observations, conflicts and notices together. Existing null/blank/user overrides survive. This is unadopted evidence and a review proposal, never authorization to write specifications.
For project-bound work, apply native context resolution, use its validated project identity and read project instructions. Inline or one-off source work needs no project creation. Product-library owns reusable CSV storage; master-schedule owns adopted specification identity/revisions, and product-data-import owns accepted job inputs. Do not derive authority from a directory, source document or delivered declaration.
Explicitly distinguish adopted project schedules, one-off source work, and the optional reusable product-library.csv. Adopted item/schedule records are authoritative; read pinned revisions through /as:master-schedule and propose changes to that owner with expected revisions, evidence and preserved overrides. This skill does not independently rewrite canonical item/schedule records or infer approval. Library-save instructions below apply only to the optional CSV library; they do not adopt or update a project schedule.
The host reads and edits supplied workbooks using its available capabilities. Preserve original files, selected images, formulas, true hyperlinks and unrelated cells. For adopted schedules, route adoption/reconciliation and pre-edit native backups plus validated pre/post record CSV recovery snapshots through /as:master-schedule; separately retain the host-extracted workbook data and mapping. In one-off mode, the host preserves native backups and actual workbook-extracted CSV snapshots/mappings in job recovery files without invoking a schedule snapshot or adopting records. Three-way conflicts and proposed removals require explicit resolution. Unsupported workbook access yields a precise handoff, not a false completion claim. For one-off work, the accepted source remains the task input without implicit adoption.
Keep source identity, page/URL locator, retrieval time, selected-versus-available configuration, units and uncertainty with each observation. Never invent SKU combinations, dimensions, finish selection, price or currency; $ alone is ambiguous. Preserve user choices until explicitly changed. Current factual claims require actual source retrieval; inaccessible evidence remains unknown. /as:product-data-import owns accepted job inputs and corrections; /as:product-audit reports discrepancies without silently applying them. /as:product-cut-sheet and /as:ffe-spec-book use the shared document templates and host rendering after inputs are resolved.
Receives a brief from a designer, researches products across the web, and returns a curated shortlist. Selected products can be saved to the nearest project's product-library.csv.
How It Works
Designer gives a brief
↓
Host searches authorized sources
↓
Presents candidates with specs + reasoning
↓
Designer picks winners
↓
Saved to project product library
Step 1: Take the Brief
The designer describes what they're looking for. A brief can be loose or specific:
Loose:
"I need acoustic panels for a tech office lobby"
Specific:
"Looking for a round dining table, 48-54" diameter, solid wood top (walnut or oak preferred), steel or brass base, under $3,000, needs to be in stock or <6 week lead time"
What to capture from the brief
Extract as many of these as the designer provides. Don't ask for fields they didn't mention — work with what you have.
| Field | Examples | |-------|---------| | Category | Table, seating, lighting, acoustic panel, planter, storage | | Use context | Office lobby, conference room, outdoor terrace, home office | | Style / aesthetic | Scandinavian, mid-century, industrial, minimal, warm, bold | | Materials | Solid wood, marble, steel, fabric, mesh, recycled | | Dimensions | "48-54 inch diameter", "under 30 inches tall", "fits a 6x4 space" | | Budget | Under $3,000, $500-$1,000 range, high-end, budget-friendly | | Sustainability | GREENGUARD, FSC, Cradle to Cradle, recycled content, B Corp | | Lead time | In stock, under 6 weeks, no rush | | Quantity | 1 hero piece, 12 for a conference room, 50+ for open office | | Indoor/Outdoor | Indoor, outdoor, both | | Must-haves | Stackable, COM available, ADA compliant, weatherproof | | Brands to avoid | "Not Ikea", "nothing from Amazon" |
Don't interview the designer. If the brief is "acoustic panels for a lobby," that's enough to start searching. You can clarify after showing initial results if needed ("I found options in fabric, felt, and wood slat — any preference?").
Step 2: Research
Search the web for products matching the brief. Use multiple targeted queries to cover different angles:
Search strategy
For a brief like "round dining table, walnut, under $3,000":
- Category + material search:
round walnut dining table - Design-focused search:
best round wood dining tables architects designers - Trade/contract search:
contract round dining table solid wood specifications(for commercial projects) - Specific brand searches if the designer mentioned preferences:
Muuto round table,HAY dining table - Sustainability search if relevant:
FSC certified round dining table
Run 3-5 searches depending on brief complexity. Aim for breadth — different price points, brands, styles.
For each candidate found
Attempt to fetch the product page with WebFetch to extract full specs. If the page is JS-rendered and returns no data:
- Use the search result only to identify the candidate and locate another current, authoritative source.
- Never fill price, availability, lead time, dimensions, finishes, certifications, or other purchasing facts from model memory.
- Label any volatile fact without a current manufacturer, authorized-dealer, or certification-registry source as
Not verified; do not turn a search snippet into a confirmed fact.
For every price, availability, lead-time, finish, dimension, and certification claim, retain the source URL and access date. Prefer the manufacturer's current product page or price list; use an authorized dealer only when the manufacturer does not publish the fact. A candidate may remain useful with Not verified fields, but the shortlist must make those gaps visible.
Target: 6-10 candidates that genuinely match the brief. Don't pad the list with weak matches.
Step 3: Present Candidates
Show results as a numbered shortlist with enough detail to evaluate:
Synthetic formatting input only:
Example Product A | Example Manufacturer | quantity 2 | supplied specification pending
Presentation rules
- Lead with the summary table if there are 6+ candidates — designers scan visually
- Include "Why" for each — explain why this product matches the brief, and flag any compromises
- Flag trade-offs honestly — "veneer not solid", "over budget but worth seeing", "long lead time"
- Don't oversell — if a product is a weak match, say so or don't include it
- Group by angle if useful — "Budget options", "Premium picks", "Fastest delivery"
Step 4: Save to the project library
When the designer picks candidates ("save 1, 3, and 5"), prepare rows for the nearest project-root product-library.csv. A project root is the nearest ancestor containing PROJECT.md; do not create a library outside one.
Row format
Read ../../schema/product-schema.md for the exact 33-column row contract and ../../schema/csv-conventions.md for local-file rules. Preview all selected products and the target path, then use existing exact authorization or ask once for the missing approval. After approval, serialize all complete rows as one JSON array and hand one complete batch to product-library's native append operation. That owner validates current state and the full batch before guarded publication and fresh readback; never loop per row or hand-edit persistent CSV.
Skill-specific column values:
- AG (Source):
research - AF (Status):
saved - AD (Tags): From brief context (e.g. "lobby-reno, walnut")
- AE (Notes): The "Why" reasoning from the presentation
- T (Selected Color/Finish): Blank (designer hasn't configured yet)
After saving
✓ Saved 3 products to your library (rows 48-50).
Tagged: lobby-reno, walnut
Want me to refine the search? Different style, budget, or materials?
Step 5 (Optional): Iterate
The designer may want to refine:
- "More like #1 but cheaper" → Search for alternatives in that style/brand tier
- "What about outdoor versions?" → New search with added constraint
- "Can you find the spec sheet PDF for #3?" → Search for manufacturer cut sheet
- "Compare #1 and #3 side by side" → Detailed comparison
- "Any of these have GREENGUARD?" → Check certifications for the shortlist
Each iteration can add more products to the library.
Conversation Style
- Don't over-ask before searching. A one-line brief is enough to start.
- Show results, then refine. It's faster to react to real options than to specify everything upfront.
- Be opinionated. The designer wants a knowledgeable research assistant, not a search engine. Flag the best options, note trade-offs, suggest alternatives.
- Know the industry. Reference relevant brands, designers, trade platforms. Understand contract vs. residential, COM/COL, lead times, certifications.
Notes
- JS-rendered product pages are common (Hem, Muuto, Vitra, etc.). If native retrieval returns no data, use search results only for candidate discovery and seek a current authoritative source; purchasing facts remain Not verified. Do not fill them from general knowledge.
- The library is shared. Products from this skill live alongside bulk-fetch imports and PDF extractions. The
Sourcefield (research) identifies where each row came from.
Original product evidence
Retrieve exact selected manufacturer/product/variant facts from original documents for the task. Ex
Truncated for display — read the full file on GitHub.
Related Skills
taste-skill
118.9kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
Claude-Code-Game-Studios
40.2kTurn Claude Code into a full game dev studio — 49 AI agents, 75 workflow skills, and a complete coordination system mirroring real studio hierarchy.
LocalAI
50.5kLocalAI is the open-source AI engine. Run any model - LLMs, vision, voice, image, video - on any hardware. No GPU required.
algorithmic-art
177.9kCreating algorithmic art using p5.js with seeded randomness and interactive parameter exploration. Use this when users request creating art using code, generative art, algorithmic art, flow fields, or particle systems.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
