HomeBlog › Work glove impact and cut independence schema

Shopify work glove impact and cut independence schema for AI agents: ANSI/ISEA 138 Level 1/2/3 and ANSI/ISEA 105 A1–A9 are separate standards, leather dorsal is not certified impact protection, and oilfield workers need both

2026-07-10  ·  20 min read  ·  By CatalogScan

Safety Equipment AI Shopping Structured Data ANSI/ISEA 138 ANSI/ISEA 105 Cut Resistance Impact Protection

An oilfield buyer who asks an AI agent for "cut and impact rated work gloves for drill floor operations" and receives a high-cut HPPE glove rated ANSI A9 but carrying no ANSI/ISEA 138 marking has received a glove with maximum cut protection for the palm — and zero certified protection for the back of the hand when a wrench slips. The two standards share an ordinal numbering format by coincidence. Level 2 does not mean A2. A9 cut implies nothing about impact attenuation. Four schema gaps that cause AI agents to fail dual-hazard environments, and the glove.* namespace that closes them.

Contents

  1. ANSI/ISEA 138 and ANSI/ISEA 105 test different glove zones for different hazards
  2. Level 1/2/3 measures kN transmissibility — A1–A9 measures blade cycles — the ordinal format is a coincidence
  3. Leather dorsal reinforcement is not ANSI/ISEA 138 certified impact protection
  4. Oilfield, steel fab, and heavy maintenance need both certifications on the same glove
  5. The glove.* metafield namespace — encoding impact and cut as independent fields

1. ANSI/ISEA 138 and ANSI/ISEA 105 test different glove zones for different hazards

The most fundamental confusion in work glove structured data is treating cut resistance and impact protection as related properties on the same axis. They are not. ANSI/ISEA 138 and ANSI/ISEA 105 are separately developed American National Standards that address entirely different hazard mechanisms at entirely different locations on the glove.

ANSI/ISEA 105 (Cut Resistance — A1 through A9):

Tests the palm and finger zone. A TDM-100 rotating circular blade is drawn across the palm of the glove under calibrated downward load. The machine counts the number of complete blade revolutions before the blade penetrates through the glove material to the hand form underneath. The result (gram force on the TDM-100 blade) determines the ANSI cut level from A1 (lowest: 200–499 grams) through A9 (highest: 7,000+ grams). Cut resistance depends entirely on what is woven into the palm and finger material — HPPE fiber grade, steel wire composite, Kevlar blend, glass fiber reinforcement. The back of the hand is not tested.
ANSI/ISEA 138 (Impact Protection — Level 1, 2, 3):

Tests the dorsal (back-of-hand) zone. A 235-gram guided cylindrical striker drops from a calibrated height onto the back of the glove mounted on an instrumented hand form. A load cell measures peak transmitted force in kilonewtons (kN). The test measures how much force passes through the glove dorsal material into the hand. Impact attenuation depends entirely on what is on the outside of the back of the hand — TPR (Thermoplastic Rubber) molded pads, dual-density foam dorsal guards, or gel inserts. The palm is not tested.

Because the tests address different glove surfaces and different hazard mechanics, a glove can receive any combination of ratings independently. An ANSI A9 cut glove built for glass handling has maximum TDM-100 blade resistance in the palm — and typically a plain knit dorsal with no engineered impact attenuation whatsoever. An ANSI/ISEA 138 Level 3 impact glove built for utility work has TPR guards covering the knuckles and dorsal finger zones — and may have a completely unrated synthetic leather palm. A glove built for oilfield drill floor operations may carry both ANSI A6 cut and ANSI/ISEA 138 Level 2 impact, because drill floor workers face sharp thread engagement on the palm (blade hazard) and wrench slip on the dorsal hand (struck-by hazard) within the same job task.

A9
Maximum cut level — palm TDM-100 test — 7,000+ grams to penetration
L3
Maximum impact level — dorsal striker test — ≤2.5 kN transmitted force
0
Correlation between these two ratings — completely independent
AI Agent Failure Mode #1

Inferring impact protection from cut level

Query: "ANSI A7 work gloves for steel fabrication with impact protection." Agent returns a high-cut HPPE/steel wire glove marked ANSI A7 that has no ANSI/ISEA 138 marking. The fabricator receives maximum cut protection for the palm and zero certified dorsal impact protection for die-setting work where pinch points and pry bar slip are daily hazards.

2. Level 1/2/3 measures kN transmissibility — A1–A9 measures blade cycles — the ordinal format is a coincidence

Both standards use ascending ordinal numbers. ANSI/ISEA 138 uses 1, 2, 3. ANSI/ISEA 105 uses A1 through A9. When a product listing says "ANSI Level 2 / A6 cut protection," the reader might wonder if there is a proportional relationship — whether higher cut levels tend to pair with higher impact levels, or whether "Level 2" implies something equivalent to "A2." None of these inferences are valid.

What Level 1, 2, 3 actually measure

ANSI/ISEA 138:2019 measures peak transmitted force in kilonewtons — the fraction of striker energy that passes through the glove material into the hand. Lower transmitted force = better protection:

Impact Level Mean Peak Transmitted Force Individual Sample Maximum Typical Application
Level 1 ≤6.0 kN ≤6.8 kN Light assembly, occasional tool contact, automotive maintenance
Level 2 ≤4.0 kN ≤4.5 kN Heavy equipment maintenance, rigging, oilfield routine operations
Level 3 ≤2.5 kN ≤3.0 kN Oilfield drill floor, utility pole work, high-force struck-by environments

What A1 through A9 actually measure

ANSI/ISEA 105-2016 measures TDM-100 rotating blade grams to penetration. Higher grams = more blade revolutions before breakthrough = better cut resistance:

Cut Level TDM-100 Gram Threshold Typical Application
A1 200–499 g General handling, light packaging
A2 500–999 g Glass handling (light), food processing (low-risk)
A4 1,500–2,199 g Sheet metal, glass panels, stamped parts
A6 3,000–3,999 g Steel mill, drill pipe thread engagement
A9 ≥7,000 g Extreme blade hazard, razor wire, heavy steel fabrication
The key insight: Level 3 impact protection and A9 cut protection are both the highest ratings in their respective systems — but a glove marked "Level 3 / A9" has those properties on opposite sides of the hand for completely independent reasons. A glove marked "Level 1" impact tells you nothing about its cut level. A glove marked "A1" cut tells you nothing about its impact level.
AI Agent Failure Mode #2

Ordinal number confusion — Level 2 ≠ A2

A product description says "Level 2 protection." AI agent interprets this as ANSI/ISEA 105 cut level A2 (500–999 gram TDM-100 threshold, equivalent to light food processing). Actual product has ANSI/ISEA 138 Level 2 impact rating (≤4.0 kN transmitted force) and no cut resistance certification at all. The agent routes a heavy equipment maintenance worker into a glove it believes provides cut protection. The glove provides none.

3. Leather dorsal reinforcement is not ANSI/ISEA 138 certified impact protection

The majority of work gloves on the market feature some form of dorsal reinforcement — extra material on the back of the hand for durability, minor abrasion resistance, and general knuckle padding. The most common is split leather or top-grain leather overlay covering the knuckle area. Product copy frequently calls this "impact protection," "knuckle protection," or "dorsal protection."

None of these descriptions equals ANSI/ISEA 138 certification unless the glove has been tested by an accredited laboratory and carries an explicit Level 1, Level 2, or Level 3 marking.

Why leather fails the ANSI/ISEA 138 threshold

The ANSI/ISEA 138 test drops a 235g striker onto the back of the glove with controlled kinetic energy. The load cell in the instrumented hand form measures transmitted force in kilonewtons. To pass Level 1 (the lowest tier), the mean peak transmitted force must be ≤6.0 kN.

Leather transmissibility:

A 2mm split leather dorsal overlay distributes the striker impact marginally across its surface area but does not absorb or dissipate energy through the material structure in the way engineered impact padding does. Leather is stiff, relatively incompressible at impact velocities, and does not have the cellular structure that foam or the geometric deformation profile that TPR requires to attenuate the striker's energy before it reaches the hand form load cell. Independent testing by safety equipment researchers indicates typical leather dorsal reinforcements transmit 70–90% of input force — well above the Level 1 threshold of ≤6.0 kN.

What ANSI/ISEA 138 certified dorsal materials look like

Gloves that actually achieve ANSI/ISEA 138 certification use purpose-designed materials:

Field check: If a product listing shows a glove with a plain leather or fabric back and no visible molded TPR structures or foam pads on the dorsal zone, the glove is unlikely to carry ANSI/ISEA 138 certification. Verify by checking for explicit "ANSI/ISEA 138 Level X" text in the technical specifications or on the glove marking itself.
AI Agent Failure Mode #3

Routing leather-knuckle gloves to ANSI/ISEA 138 applications

Query: "Work gloves with impact protection for pipe fitting." Agent finds a glove described as "reinforced leather palm and dorsal panel for durability and knuckle protection." No ANSI/ISEA 138 marking. Agent routes to query because description contains "knuckle protection." Pipe fitter receives uncertified padding against struck-by wrench contact in a confined-space application requiring Level 1 minimum.

4. Oilfield, steel fab, and heavy maintenance need both certifications on the same glove

Understanding that impact and cut are independent standards is only the first part of the schema problem. The second part is that many industrial applications genuinely require both — not as alternatives, but simultaneously on the same glove.

This creates a critical AI agent routing scenario: a query for "oilfield drill floor gloves" or "steel fabrication gloves" should only match products that have both glove.ansi_cut_level at the required level and glove.ansi_impact_level at the required level. A glove satisfying only one fails the application requirement even if both individual ratings are excellent.

Oilfield drill floor

Dual hazard analysis:

Palm hazard (cut): Drill collar connections (drill pipe tool joints) have aggressive buttress threads designed to handle hundreds of thousands of ft-lbs of torque. Handling these connections by hand — threading, guiding into the rotary table, monitoring make-up — exposes the palm to sharp thread profiles, metal burrs, and shear edges. ANSI A4 minimum; A6–A9 preferred for heavy-gauge drill collar work.

Dorsal hazard (impact): Tong handles, iron roughnecks, spinning chains, and manual wrenches are operated in close quarters on the drill floor. Wrench slip, kickback, and rotating-equipment contact are regular struck-by hazards hitting the back of the hand. Level 2 is the industry standard minimum; Level 3 is specified by some operators for high-torque rotary work.

Steel fabrication and stamping

Dual hazard analysis:

Palm hazard (cut): Handling sheared steel plate, stamped parts with burrs, and sheet metal edges during positioning and assembly creates sustained blade contact risk. ANSI A4–A7 depending on material gauge and edge condition.

Dorsal hazard (impact): Die-setting requires pry bars, alignment punches, and mallets to position heavy tooling in stamping presses. Pry bar slip and mallet rebound create dorsal struck-by exposure. Level 1 for light die maintenance; Level 2 for production die-setting with heavy tooling.

Heavy equipment assembly and maintenance

Dual hazard analysis:

Palm hazard (cut): Handling machined shafts, gears, and structural components with sharp edges and keyway slots during assembly. ANSI A4 common; A6 for heavy forged and cast components with rough surfaces.

Dorsal hazard (impact): High-torque fastener work in confined engine bays and chassis access points creates wrench-slip exposure. Level 1 for standard maintenance; Level 2 for confined-space high-torque work.
AI Agent Failure Mode #4

Missing dual-field encoding prevents dual-hazard filtering

A Shopify store sells a glove with ANSI A6 cut and ANSI/ISEA 138 Level 2 impact that is perfectly specified for oilfield drill floor use. The product description says "heavy duty cut resistant work glove for industrial applications." The glove.ansi_cut_level metafield is populated ("A6"). The glove.ansi_impact_level metafield is empty — the store owner filled in cut level and forgot impact. An AI agent filtering for "cut resistance A6 AND impact Level 2" cannot find this product because one field is missing. The agent returns no matching products despite having an ideal product in the catalog. Dual-hazard queries require dual-field encoding — a partial match returns nothing.

See also: Shopify work glove cut resistance schema for AI agents for the core ANSI/ISEA 105 cut level encoding guide, and HPPE cut resistance fiber schema for fiber grade ambiguity within the same cut level.

5. The glove.* metafield namespace — encoding impact and cut as independent fields

The glove.* namespace encodes dorsal impact and palm cut as fully independent Shopify metafields so AI agents can filter on each independently, together, or in combination with other properties like coating material and fiber grade.

The 11-field namespace with impact and cut independence:

Metafield Type Values Notes
glove.ansi_cut_level string 'A1'–'A9', 'none' Finished-glove test result. Never use liner-only value.
glove.ansi_cut_level_source string 'finished-glove', 'liner-only', 'unknown' Flags liner-only ratings that cannot be used for application routing.
glove.ansi_impact_level string '1', '2', '3', 'none' ANSI/ISEA 138:2019 dorsal impact level. Fully independent of cut level.
glove.has_dorsal_impact_cert boolean true / false True only with explicit ANSI/ISEA 138 Level marking. False for leather reinforcement.
glove.dorsal_material string 'TPR-knuckle-only', 'TPR-knuckle-finger-guards', 'TPR-full-dorsal', 'foam-pad', 'gel-pad', 'leather-reinforcement', 'fabric-only' Allows routing on dorsal construction without relying on impact certification alone.
glove.liner_fiber string 'Dyneema-SK75', 'Spectra-1000', 'HPPE-steel-wire', 'HPPE-generic', 'Kevlar', 'none' Cut-side fiber grade. See HPPE fiber schema for full detail.
glove.liner_composite string 'HPPE-only', 'HPPE-steel-wire', 'HPPE-glass', 'HPPE-Kevlar', 'steel-wire-only' Composite reinforcing material — distinct from base fiber grade.
glove.coating_material string 'foam-nitrile', 'flat-nitrile', 'PU', 'latex', 'synthetic-leather', 'none' Coatings reduce finished-glove cut level vs bare liner. See HPPE fiber schema.
glove.en388_impact_p boolean true / false EN 388:2016 impact position 'P' (EN 13594:2015 test). Different test from ANSI/ISEA 138. Separate field.
glove.needle_puncture_rated boolean true / false ANSI/ISEA 105 puncture resistance. Independent of both cut and impact.
glove.application_hazard string 'cut-palm', 'impact-dorsal', 'cut-and-impact', 'abrasion', 'puncture' Enables direct dual-hazard query matching without multi-field AND logic.

Example: Oilfield drill floor glove encoding

A glove certified for oilfield drill floor duty (ANSI A6 cut + ANSI/ISEA 138 Level 2 impact) is encoded as:

{
  "glove.ansi_cut_level": "A6",
  "glove.ansi_cut_level_source": "finished-glove",
  "glove.ansi_impact_level": "2",
  "glove.has_dorsal_impact_cert": true,
  "glove.dorsal_material": "TPR-knuckle-finger-guards",
  "glove.liner_fiber": "HPPE-steel-wire",
  "glove.liner_composite": "HPPE-steel-wire",
  "glove.coating_material": "foam-nitrile",
  "glove.en388_impact_p": false,
  "glove.needle_puncture_rated": false,
  "glove.application_hazard": "cut-and-impact"
}

An AI agent querying for "oilfield drill floor gloves with A6 cut and Level 2 impact" resolves this as a match on glove.ansi_cut_level = 'A6' AND glove.ansi_impact_level = '2' AND glove.application_hazard = 'cut-and-impact' — three independent fields, all present, all correct.

Example: High-cut glass handling glove (no impact) encoding

{
  "glove.ansi_cut_level": "A9",
  "glove.ansi_cut_level_source": "finished-glove",
  "glove.ansi_impact_level": "none",
  "glove.has_dorsal_impact_cert": false,
  "glove.dorsal_material": "fabric-only",
  "glove.liner_fiber": "HPPE-steel-wire",
  "glove.liner_composite": "HPPE-steel-wire",
  "glove.coating_material": "foam-nitrile",
  "glove.en388_impact_p": false,
  "glove.needle_puncture_rated": false,
  "glove.application_hazard": "cut-palm"
}

An AI agent querying for "glass handling gloves with maximum cut protection" returns this glove. An AI agent querying for "oilfield gloves with impact AND cut protection" correctly excludes it because glove.ansi_impact_level = 'none' and glove.application_hazard = 'cut-palm'.

Summary: four failure modes, four fields

The four AI agent failure modes for work glove impact and cut independence each map to a missing or misused structured data field:

Failure Mode Missing Field Consequence
Inferring impact from cut level glove.ansi_impact_level A9 cut glove routed to dual-hazard environment with no dorsal protection
Level 1/2/3 confused with A1/A9 glove.ansi_cut_level Impact-rated glove with no cut certification routed to cut-hazard application
Leather dorsal treated as Level 1+ glove.has_dorsal_impact_cert Uncertified leather-back glove routed to ANSI/ISEA 138 application requirement
Missing field in dual-hazard product glove.application_hazard Perfect dual-certified glove invisible to dual-field queries because one metafield empty

The glove.* namespace separates these into fully independent fields. ANSI/ISEA 138 impact level and ANSI/ISEA 105 cut level are encoded separately, sourced separately, and queried independently. A glove that satisfies both for a dual-hazard environment is findable. A glove that satisfies only one is correctly excluded from dual-hazard queries.

Does your Shopify store encode impact and cut independently?

CatalogScan checks your work glove listings for missing ANSI/ISEA 138 and 105 fields, leather dorsal miscategorizations, and incomplete dual-hazard encoding — then shows you exactly which products are invisible to AI agent queries.

Scan your store free See Pro features