ZooData β Competitor Intelligence Monitor
Know your enemy. Two modes: Full Scan + Quick Check. Respond in user's language.
Files
| File | Purpose |
|---|
{skill_base_dir}/βscripts/βzoodata.py | Execute for all API calls (run --help for params) |
{skill_base_dir}/βreferences/βreference.md | Load for exact field names or response structure |
{skill_base_dir}/βmonitor-data/ | Runtime storage (auto-created): config.json, baseline.json, history/, alerts.json |
Credential
Required: ZOODATA_API_KEY. Get free key at zoodata.ai/βapi-keys.
Capabilities & Data Flow
- Network: only
https://api.zoodata.ai (Bearer ZOODATA_API_KEY). Setting ZOODATA_BASE_URL to any other host triggers a CLI warning β do not point it at hosts you don't trust.
- Execution: bundled shared ZooData CLI
{skill_base_dir}/βscripts/βzoodata.py (Python 3, stdlib-only). The shared CLI exposes ALL ZooData endpoints as subcommands; this skill's workflows use: categories, market, competitors, products, product, history, analyze, competitor-analysis, check, plus the review fallback toolkit (reviews-raw / review-tag-prompt / review-reduce-prompt / review-aggregate). Do not invoke unrelated subcommands for this skill's tasks.
- Local files: a temporary
/βtmp/βreview_<ASIN>_<timestamp>/ working dir during the review fallback; reads the optional credential store ~/β.zoodata/βconfig.json.
- Sent to the API: keywords, category paths, ASINs, marketplace/βdate and numeric filter values only. Never sent: budget, experience level, risk tolerance, or any other user-profile text β profile inputs map client-side to numeric filters.
- Credits: every API call consumes account credits. For broad or ambiguous requests, state the estimated credit cost and confirm with the user before running multi-call scans. The composite
competitor-analysis command executes ~17+ API calls (Full Scan, ~28-35 credits documented) in ONE invocation and has NO skip/βtrim flags β under a credit cap, use the granular commands instead.
Input
Required: keyword or ASIN(s). Optional: my_asin, competitor_asins, brand.
If only ASIN given β derive keyword via product --asin then ask user to confirm.
Brand queries MUST also include confirmed --category.
API Pitfalls (CRITICAL)
- Category auto-detection: categoryPath is auto-detected from keyword, ASIN, or top search result. If
category_source in output is inferred_from_search, MUST confirm with user before trusting results
- All keyword-based endpoints MUST include
--category; ASIN-specific endpoints do NOT need it
- Brand + category: a brand sells across categories β only analyze within locked subcategory
- Use API fields directly: revenue=
sampleAvgMonthlyRevenue (NEVER priceΓsales), sales=monthlySalesFloor, concentration=sampleTop10BrandSalesRate
- reviews/βanalysis: needs 50+ reviews. Fallback chain when sample is insufficient:
- Lightweight:
realtime/βproduct ratingBreakdown β only star distribution, no themes
- Full 11-dim insights β bypass
/βreviews/βanalysis entirely:
a. zoodata.py reviews-raw --asin X β fetch up to 100 raw reviews (10 credits, ~60s)
b. For each review: render Map prompt via zoodata.py review-tag-prompt --review '<json>'
and have your own LLM produce JSON tags (sentiment + 11 dimensions)
c. Collect candidate phrases per dimension; for each dimension render
Reduce prompt via zoodata.py review-reduce-prompt --label-type X --candidates '[...]'
and have your LLM produce semantic clusters
d. zoodata.py review-aggregate --reviews R --tagged T --clusters C
β consumerInsights output compatible with /βreviews/βanalysis
- Fallback caveats (apply to the 4-step chain above β lessons from end-to-end validation):
- Working dir:
WORK=/βtmp/βreview_<ASIN>_$(date +%s) && mkdir -p $WORK
- Step b CLI behavior:
review-tag-prompt RENDERS the prompt only; YOUR LLM produces the JSON. Render once to learn the schema, then produce tags for all N reviews in one in-context pass (don't call the CLI N times).
- Step c candidate extraction (Python one-liner):
candidates = {d: sorted({el.strip().lower() for t in tagged for el in (t.get(d) or [])}) for d in DIMS}
- Small-sample rule (reviewCount<50): demote single-mention items πβπ; NEVER attach table-level or section-header π when any row inside is π; suppress "π΄ Critical" verdicts on count=1
- Scope: fallback replaces ONLY the
/βreviews/βanalysis aggregation. This skill's primary workflow outputs (competitor metrics, brand ranking, pricing, etc.) remain valid β do not re-run them.
On Missing Key
When ZOODATA_API_KEY is not set (verify via python {skill_base_dir}/βscripts/βzoodata.py check β exits 2 if no key in env or ~/β.zoodata/βconfig.json): follow the "On Missing Key" protocol in zoodata/βSKILL.md β STOP before any call, link the user to https://zoodata.ai/en/api-keys, and DO NOT produce a "partial analysis from public knowledge" / "for reference only" fallback as a substitute.
On 401 Invalid Key
When zoodata.py returns code 401: follow the "On 401 Invalid Key" protocol in zoodata/βSKILL.md β STOP further calls, tell the user the key was rejected and direct them to api-keys, do not fabricate missing data.
On 402 Credit Exhausted
When zoodata.py returns code 402: follow the "On 402 Credit Exhausted" protocol in zoodata/βSKILL.md β STOP further calls, report partial findings already gathered, do not fabricate missing data.
Mode Selection
- Full Scan (~28-35 credits): First run, no baseline.json, explicit request, or weekly refresh
- Quick Check (~5-10 credits): Cron trigger, baseline exists, "check competitors"
Full Scan Flow
competitor-analysis --keyword X [--category Y] [--my-asin Z] (composite, auto-detects category)
- If
category_source is inferred_from_search, confirm with user before presenting results
- Analyze & score β save baseline to
{skill_base_dir}/βmonitor-data/ β offer Auto-Monitor
Quick Check Flow
- Load config.json + baseline.json from
{skill_base_dir}/βmonitor-data/ (missing β fall back to Full Scan)
- Poll
product --asin {asin} for each tracked ASIN
- Diff against baseline with tiered alerts β update baseline β offer Auto-Monitor
Alert Tiers
| π΄ Critical | π‘ Watch | π’ Opportunity |
|---|
| Price change > threshold | FBAβFBM switch | Competitor stock-out |
| BSR crash > threshold | Rating change | Bullet/βimage changes |
| Buy Box owner changed | Abnormal review growth | Variant added/βremoved |
| Title modified | |
Competitive Score (per competitor, 1-100)
| Dimension | Weight | 80-100 (Strong) | 50-79 (Moderate) | 0-49 (Weak) |
|---|
| Sales Dominance | 25% | Top 3 in category, >5K units/βmo π | Top 20, 1K-5K units/βmo π | Below Top 20, <1K units/βmo π |
| Brand Strength | 20% | Brand in CR10, 5+ SKUs, wide price range π | Known brand, 2-4 SKUs π | Unknown brand, single SKU π |
| Listing Quality | 20% | 7+ images, 5 bullets, A+, optimized title π | 5-6 images, basic bullets π | <5 images, weak bullets, no A+ π |
| Customer Satisfaction | 20% | Rating β₯4.5, <3% 1-star, positive sentiment π | 4.0-4.4, 3-8% 1-star π | <4.0 or >8% 1-star π |
| Trend Momentum | 15% | BSR improving 30d, sales growth >10% π | BSR stable, flat sales π | BSR declining, sales drop π |
Competitive Threat Level
| Total Score | Threat | Interpretation |
|---|
| 80-100 | π΄ Dominant | Hard to compete head-on; find differentiation or avoid price band π‘ |
| 50-79 | π‘ Competitive | Beatable with better listing, pricing, or reviews π‘ |
| 0-49 | π’ Vulnerable | Weak competitor; opportunity to capture share π‘ |
Market Structure Analysis
- CR10 > 70%: Concentrated market β new entrants need strong differentiation or niche positioning π
- CR10 40-70%: Moderately competitive β room for well-positioned products π
- CR10 < 40%: Fragmented β opportunity for brand building π
- Top brand share > 25%: Category leader dominance β avoid direct competition in their price band π‘
- New SKU rate > 15%: Active market with frequent new entrants π
- New SKU rate < 5%: Mature/βstagnant market, high barriers π
Auto-Monitor Prompt
After EVERY run, offer: "Set up automatic monitoring? I can generate a scheduled Quick Check." Provide platform-specific setup (OpenClaw /βcron, ChatGPT Scheduled Tasks, Claude Projects).
Output Spec
Full Scan sections: Battlefield Overview β Competitor Matrix β Brand Power Ranking β Price Map β 30-Day Trends β Review Battle β Listing Audit β Competitive Scores β Battle Strategy β Data Provenance β API Usage.
Language (required)
Output language MUST match the user's input language. If the user asks in Chinese, the entire report is in Chinese. If in English, output in English. Exception: API field names (e.g. monthlySalesFloor, categoryPath), endpoint names, technical terms (e.g. ASIN, BSR, CR10, FBA, credits) remain in English.
Disclaimer (required, at the top of every report)
Data is based on ZooData API sampling as of [date]. Monthly sales (monthlySalesFloor) are lower-bound estimates. This analysis is for reference only and should not be the sole basis for business decisions. Validate with additional sources before acting.
Confidence Labels (required, tag EVERY conclusion)
- π Data-backed β direct API data (e.g. "CR10 = 54.8% π")
- π Inferred β logical reasoning from data (e.g. "brand concentration is moderate π")
- π‘ Directional β suggestions, predictions, strategy (e.g. "consider entering $10-15 band π‘")
Rules: Strategy recommendations are NEVER π. Anomalies (>200% growth) are always π‘. User criteria override AI judgment.
Aggregate-label rule (applies to ALL report output, not just fallback): NEVER attach π to ANY element that aggregates or groups underlying content when ANY piece of that content is π or π‘. "Aggregate/βgrouping elements" include:
- Section headers at EVERY level (
#, ##, ###, ####) β including top-level summary sections like "Overall Score", "Verdict", "Executive Summary"
- Summary/βscore lines anywhere in the report (e.g.
## Overall Score β 27/β100 Β· Grade F π is WRONG if any Basis row inside is π)
- Table column headers in comparison tables (e.g.
**Target ASIN** π as a column label is WRONG if any cell in that column contains π)
- Table row headers or row-aggregation labels (when the row aggregates multiple cells of mixed confidence)
- Any other visual grouping label β bullet-list group titles, callout box titles, etc.
A group-level π implies the whole block/βcolumn/βrow is data-backed, which smuggles inferred/βdirectional content into the π tier via visual grouping. Either (a) omit the group-level label entirely (preferred when content mixes tiers), or (b) use the LOWEST confidence present inside (π if any underlying content is π; π‘ if any is π‘). This is a universal output-quality rule β it applies regardless of which fallback path (if any) was triggered.
Emoji reservation rule (closely related): The three confidence symbols π π π‘ are RESERVED for confidence labeling. NEVER use them as decorative prefixes on section headers, table headers, or any aggregate element β even when you also include a correct confidence suffix on the same line. Example:
- β WRONG:
## π Overall Score β 27/β100 Β· Grade F π (the leading π reads as a data-backed claim even though the trailing π is correct)
- β
RIGHT:
## Overall Score β 27/β100 Β· Grade F π (no decorative emoji, just the proper confidence suffix)
- β
RIGHT:
## π― Overall Score β 27/β100 Β· Grade F π (use non-reserved decorative icons like π― π§ π π π π π¨ π π when a visual prefix is desired)
Decorative emoji β confidence label β but from a reader's perspective, a leading π/π/π‘ is indistinguishable from a confidence claim. Reserve these three symbols EXCLUSIVELY for confidence annotation to avoid ambiguity.
Data Provenance (required)
Include a table at the end of every report:
| Data | Endpoint | Key Params | Notes |
|---|
| (e.g. Market Overview) | markets/βsearch | categoryPath, topN=10 | π Top N sampling, sales are lower-bound |
| ... | ... | ... | ... |
Extract endpoint and params from _query in JSON output. Add notes: sampling method, T+1 delay, realtime vs DB, minimum review threshold, etc.
API Usage (required)
| Endpoint | Calls | Credits |
|---|
| (each endpoint used) | N | N |
| Total | N | N |
Extract from meta.creditsConsumed per response. End with Credits remaining: N.
API Budget
Full Scan: ~28-35 credits (all 11 endpoints via composite). Quick Check: ~5-10 credits (realtime/βproduct Γ N ASINs).