v-modal/vmodal_mcp / skills/vmodal-trend-radar
vmodal-trend-radar
VerifiedMeasure fashion trends over time in VModal street-style video collections: for each trend item (garment, color, print, accessory) search every season / sub-collection, verify every counted scene on real frames, and build an item × period matrix that says what is rising, fading, stable or evergreen, with proof frames. Use when the user asks "what's trending", "trend radar", "is X rising / fading", "evergreen pieces", or compares seasons or street vs brand, e.g. "Run a Milan trend radar for suede and leopard. Collection: fashion_milan". Requires the vmodal MCP and the vmodal-find skill.
| File | Kind | Size |
|---|---|---|
| SKILL.md | skill md | 7.5 KB |
| reference.md | markdown | 2.8 KB |
| scripts/fashion_period.py | script | 3.2 KB |
| scripts/trend_matrix.py | script | 4.4 KB |
| tests/test_trend_matrix.py | script | 5.7 KB |
reference.md
View on GitHub# vmodal-trend-radar reference
## Default items (12)
| # | Item | Kind | Notes for verification |
|---|---|---|---|
| 1 | barrel jeans | silhouette | curved, wide leg tapering at ankle; straight wide jeans fail |
| 2 | burgundy | color | on a main garment; wine/oxblood pass, red/brown fail |
| 3 | suede jacket | material | napped matte leather look; smooth leather fails |
| 4 | leopard print | print | on a garment/shoes; bag-only = PARTIAL |
| 5 | ballet flats | shoes | flat, rounded toe; loafers fail |
| 6 | trench coat | outerwear | belted / double-breasted, any color |
| 7 | long wool coat | outerwear | below the knee |
| 8 | sheer or mesh | material | see-through layer visible |
| 9 | boat shoes | shoes | moccasin stitching, laces at side |
| 10 | oversized blazer | tailoring | shoulders dropped or past hips |
| 11 | scarf as accessory | accessory | worn on bag/head/neck as styling, not winter muffler only |
| 12 | red accent | color | one small red item (bag, shoes, tights) in an otherwise neutral look |
## Collection map
| Collection | Column | Sub-collection |
|---|---|---|
| fashion_milan | summer | `Most_Iconic_Summer_Street_Fashion_in_Milan_` |
| fashion_milan | fall | `Fall_Fashion_Trends_and_Classic_Pieces_for_dressing_this_Autumn` |
| fashion_milan | winter | `Italian_Winter_Fashion_Best_Winter_Style_ideas_You_ll_want_to_recreate` |
| fashion_all | brand (optional) | `fashion_brand_milan` |
Columns are where you search; **periods** (matrix columns) come from video filenames,
e.g. `NOVEMBER_2025_…` → `2025-11`, `FASHIONABLE_OUTFITS_WINTER_2026_…` → `2026-winter`,
`Discover_Milan_s_June_Vibe…` → `summer-unknown` (no year: shown, never used for trends).
Check one with `python <skill_dir>/scripts/fashion_period.py period <filename> <sub_collection>`.
## Budgets
| What | Default | Deep run |
|---|---|---|
| items | ≤ 12 | ≤ 12 |
| scenes verified per cell | 6 | 8 |
| scenes per video per cell | 2 | 3 |
| pages per cell | 1 (+1 if < 6 scenes) | 3 |
| frames per bulk call | ≤ 20 | ≤ 20 |
12 items × 3 columns × 6 = 216 frames max; tell the user when a run will exceed ~100 frames
and offer to start with 4 items.
## Thresholds (in `scripts/trend_matrix.py`)
`MIN_VIEWED=4`, `MIN_PERIODS=3`, `DELTA=0.15`, `EVERGREEN=0.5`, `N_EVIDENCE=3`.
Rates are PASS / viewed among the scenes search surfaced for that item, so they compare
periods for the same item; they are not prevalence in the population.
## Worked example (shape; counts are illustrative)
`suede jacket`, fall column: 20 hits → 9 scenes → top 6 (≤ 2 per video) → frames: 4 PASS,
1 PARTIAL (jacket hidden by bag), 1 FAIL (smooth leather). The 6 lines go into
`verdicts.jsonl`; the script spreads them across `2025-09` and `2025-11` by filename.
## Tests
`python skills/vmodal-trend-radar/tests/test_trend_matrix.py test_all`
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 471k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 322k | Updated | Claude CodeCodex | Verified | Install |