compliance-check — independently scanned and version-tracked by SaferSkills.
SaferSkills independently audited compliance-check (Agent Skill) and scored it 100/100 (green). The audit ran 55 deterministic rules across Security, Supply Chain, Maintenance, Transparency, and Community; it found 0 high-severity and 0 lower-severity findings. The full rule-by-rule trace and per-finding evidence are below. Free, methodology-open.
Findings & checks · 0 flagged
Every scanned point with the score it earned and what moved between them.
First recorded scan — no prior version to compare against.
The primary manifest — the file an agent reads to learn what this artifact does.
Assess compliance likelihood for any consumer product before listing for US sale.
Usage: /compliance-check [input]
Accepted inputs:
E581D9BF-BCBC-421B-B12D-3762565A1C50, or numeric 18+ digit string)https://cjdropshipping.com/product/xxx)B08GFK757R)"bamboo cutting board with juice groove")This skill produces evidence-based assessments, not legal opinions. A 🟡 verdict means "market evidence suggests the supplier has documentation — request it." It does not mean the product is safe to list without obtaining actual certificates.
No product advances past rfq-sent status without a compliance verdict from this skill.
Determine input type:
CJ PID (UUID pattern xxxxxxxx-xxxx-xxxx-xxxx-xxxxxxxxxxxx, numeric 18+ digits, or CJHD prefix): If you have a CJ lookup script in your project, run it with --pid [PID] --pretty. Otherwise, use the CJ product URL directly or the product name as a search anchor. Extract: name, category, description, weight_g, length_cm, width_cm, height_cm, variants[].price.
CJ product URL: Extract PID from URL path, then proceed as CJ PID above.
Amazon ASIN (B0 prefix + alphanumeric, 10 chars): WebFetch https://www.amazon.com/dp/[ASIN] — extract title, description, age rating, compliance signals.
Product name / free text: Use as search anchor directly. Skip CJ API lookup.
Store these for use throughout:
PRODUCT_NAME — resolved product namePRODUCT_INPUT — original input as-isPRODUCT_CATEGORY — assigned in Step 2SEARCH_ANCHOR — best 3–5 word phrase for searches (e.g. "bamboo cutting board" not the full CJ title)Read references/compliance-matrix.md.
Identify which category applies based on product name, description, and intended use. Match at least two classification signals before committing. If a product spans two categories (e.g., an electronic toy for children), apply both sections and use the stricter requirement for any overlap.
Categories:
Set PRODUCT_CATEGORY to the matched category name(s).
Using the relevant section(s) from references/compliance-matrix.md:
Do not hallucinate compliance requirements. Use only what is in the reference file.
Read references/market-evidence-guide.md for the signal strength ranking and category-specific search channels.
Run the searches listed under the matched category in the guide. For each relevant result, WebFetch the product page.
For each listing found, extract the fields from the "What to Extract" table in the guide: listing status, compliance standard cited, certificate linked, self-claims, safety labels, brand, ASIN/SKU.
Stop early if 🟢-level evidence is found (official compliance document located).
Run the cross-category search:
"[SEARCH_ANCHOR]" "[primary standard for PRODUCT_CATEGORY]" OR "certified" OR "compliant" OR "third-party tested"Run one additional broad search without site filter to surface DTC brand sites and editorial coverage:
"[SEARCH_ANCHOR]" -site:amazon.com -site:walmart.com -site:target.comExtract the same fields as Step 4. Combine all evidence before assigning a verdict.
Based on all evidence, assign one verdict:
| Verdict | Criteria |
|---|---|
| 🟢 Strong Evidence | Official compliance document found (CPC PDF, FCC ID on fccid.io, CoA, etc.) OR active major-retailer listing with explicit standard cited |
| 🟡 Likely Compliant | Active Amazon or major retailer listing found (retailer enforcement implies documentation exists), but no direct certificate located. Product type has a well-understood compliance path. |
| 🟠 Uncertain | Limited marketplace presence (DTC-only or sparse results); compliance docs not confirmed; category is moderately complex or has multiple applicable standards |
| 🔴 High Risk | Not found on Amazon or major retailers; compliance-sensitive category; no compliance references found; or product characteristics trigger high-risk flags (e.g. loose magnets for children, consumable for infants, mains-powered without FCC ID) |
Set Confidence: High / Medium / Low based on how many sources were searched and how current the results appear.
List Key risks: 2–4 specific risk factors for this product based on its characteristics.
Read references/report-template.md.
Directory: compliance-reports/ (adjust to your project's path) Filename: YYYY-MM-DD-[product-slug].md (slug = lowercase-hyphenated product name)
Follow the template exactly. Do not leave placeholders. Add a category field in the frontmatter matching PRODUCT_CATEGORY.
If this product appears in the active product catalog, find its entry and update the compliance-docs field:
[verdict emoji] [verdict label] — [[YYYY-MM-DD-product-slug]]🟡 Likely Compliant — [[2026-03-28-bamboo-cutting-board]]Optionally log to your memory or notes system:
---
type: recall
date: YYYY-MM-DD
tags: [compliance-check, product-assessment]
---
Ran compliance assessment for [PRODUCT_NAME] (input: [PRODUCT_INPUT]).
Category: [PRODUCT_CATEGORY]. Verdict: [emoji + label] — [1-sentence evidence summary].
Key risks: [comma-separated list].
Recommended action: [top action item].
Report: compliance-reports/YYYY-MM-DD-[slug].mdcompliance-matrix.md.references/compliance-matrix.md immediately.assets/approved-examples/.~30 seconds. Free. No account. Every finding cites a rule and a line of evidence.