Produces a consultant-grade warehouse racking + MHE/pedestrian-segregation safety artifact for a named warehouse, racking installation, and traffic layout — never a generic 'inspect carefully' or 'wear hi-vis'. Use this skill to assess pallet racking, build a racking inspection regime, classify rack damage on the SEMA Red/Amber/Green bands, plan MHE/forklift and pedestrian segregation, or check a warehouse against BS EN 15635 / SEMA / OSHA 1910.178 / PUWER / HSE L117. It is design- and engineering-control-led: racking is controlled by SWL-rated configuration, column protection and the EN 15635 inspection regime (PRRS, weekly visual, >=12-monthly expert), and MHE/pedestrian conflict by ENGINEERED segregation (barriers, one-way systems, marked walkways) before hi-vis. It gets the SEMA RAG priority right (Green=monitor; Amber=repair in 4 weeks + escalate; Red=immediate off-load + isolate), refuses a vague 'a warehouse', never assumes an SWL ([GAP]). Decision-support only; a competent person must review.
Produces a consultant-grade warehouse racking + MHE/pedestrian-segregation safety artifact for a named warehouse, racking installation, and traffic layout — never a generic 'inspect carefully' or 'wear hi-vis'. Use this skill to assess pallet racking, build a racking inspection regime, classify rack damage on the SEMA Red/Amber/Green bands, plan MHE/forklift and pedestrian segregation, or check a warehouse against BS EN 15635 / SEMA / OSHA 1910.178 / PUWER / HSE L117. It is design- and engineering-control-led: racking is controlled by SWL-rated configuration, column protection and the EN 15635 inspection regime (PRRS, weekly visual, >=12-monthly expert), and MHE/pedestrian conflict by ENGINEERED segregation (barriers, one-way systems, marked walkways) before hi-vis. It gets the SEMA RAG priority right (Green=monitor; Amber=repair in 4 weeks + escalate; Red=immediate off-load + isolate), refuses a vague 'a warehouse', never assumes an SWL ([GAP]). Decision-support only; a competent person must review.
A consultant-grade, design- and engineering-control-led warehouse racking + MHE/pedestrian-
segregation safety assessment for a named warehouse, racking installation, and traffic layout —
never a generic "inspect carefully" or "wear hi-vis" statement. Its entire reason to exist is that
: racking by the
, , and the
(a named , the inspection and the inspection); and
MHE/pedestrian conflict by (physical barriers, one-way systems, marked and
protected walkways and crossings) — any reliance on hi-vis, signage, or "look out for
forklifts." A treatment that leads with "inspect carefully" / hi-vis, with no SWL design, no
inspection regime, and no engineered segregation, is (). That is the failure mode this skill exists
to prevent.
## Output format
racking and traffic risk are controlled by design and engineering first
SWL-rated configuration
column/upright protection
BS EN 15635 inspection regime
PRRS
weekly visual
≥12-monthly expert
engineered segregation
before
flagged and pushed up the hierarchy — never
accepted as the headline control
KB-SNIP-RACKING-MHE
It forces the single lever that separates a defensible artifact from copy-paste paperwork:
site/installation specificity plus the full hierarchy of controls. It applies the SEMA RAG
remedial priority to every damage finding and gets the escalation right — Green = log + monitor;
Amber = repair/replace within 4 weeks and auto-escalate to Red if not actioned; Red =
immediate off-load + isolate the affected run (a Red finding down-rated to "monitor" is rejected).
It refuses to assess "a warehouse" or "looks fine" without a named site + the racking
configuration + the traffic layout, and never assumes a site-specific SWL, tolerance, or
inspection interval — an unsupplied value is a literal [GAP], never an invented rating. Grounded in
BS EN 15635:2008 / SEMA Code of Practice, OSHA 29 CFR 1910.178, PUWER 1998, and HSE ACOP
L117 / HSG136. Decision-support only; a competent person must review the output.
When to use this skill
Use this skill when the user needs a warehouse racking + MHE/traffic safety assessment for a
concrete site, racking installation, and traffic layout — for example "assess the pallet racking and
forklift routes in the Coventry DC", "build a racking inspection regime and PRRS appointment for the
chilled warehouse", "classify the damage on bays A12–A18 against the SEMA RAG bands", or "plan
MHE/pedestrian segregation for the goods-in marshalling area". It is not for a generic "how do I
keep a warehouse safe?" answer: the Workflow intake below forces the named site, the racking
configuration, and the traffic layout before any drafting, and refuses a vague "a warehouse" / "looks
fine" request.
Data Protection & De-identification (MANDATORY — apply before drafting)
Apply this BEFORE you draft anything. Treat injury, illness, and any health
detail as the highest sensitivity. Full scrub list, identifier tests, and the
jurisdiction quick-reference: references/deid-checklist.md.
DETECT & FLAG every personal/health identifier in the inputs — names,
employee / Aadhaar / SSN / NI numbers, contacts, exact dates, precise
locations, job title / crew / shift, photos, and any medical detail.
List what you found before drafting. If unsure whether something is
identifying, treat it as identifying.
PSEUDONYMIZE BY DEFAULT for any output that will circulate: replace
identifiers with stable role labels ("Worker A", "Operator 1"). Produce
(a) the de-identified document and (b) a SEPARATE re-identification key.
Never put the key or any name↔label mapping in the document. Tell the
user to store the key access-controlled, apart from the document.
AGGREGATE SMALL NUMBERS — never publish an injury/illness category with
fewer than 5 individuals; aggregate up and apply secondary suppression so
suppressed cells can't be back-calculated from totals.
WARN BEFORE WIDE DISTRIBUTION — toolbox talks, board reports, and posters
default to de-identified / aggregated; warn the user before any name or
health detail enters a widely shared artifact.
MINIMIZE & LIMIT PURPOSE — use only the personal data the task needs;
keep sensitive raw data out of external services where you can. When in
doubt, ask before including it.
Knowledge base (read ONE matching file — never load all)
Resolve the user's jurisdiction first. Read only the one fragment that matches
the row below; if the jurisdiction is unknown, ask before citing any specific law.
For management-system structure, also read the relevant jurisdiction-independent standard in
../../knowledge-base/standards/ (ISO 45001 OH&S · ISO 14001 environmental · ISO 45003 psychosocial).
Always apply ../../knowledge-base/prompt-snippets/hierarchy-of-controls.md (KB-SNIP-HOC)
to every control recommendation. For any benchmark/figure, look up the ID in the relevant
_registry.yaml, then read ONLY the named file — and quote its source+year.
Jurisdiction / scope
Read
USA
../../knowledge-base/regulatory/mhe-pit.md (KB-REG-MHE-PIT) — OSHA 29 CFR 1910.178 powered industrial trucks: operator training/evaluation (1910.178(l)), pre-use inspection + defect removal-from-service, and HSG136/ANSI MH16.1-style engineered pedestrian-vehicle segregation BEFORE "look out for forklifts"
UK / EU
../../knowledge-base/regulatory/mhe-pit.md (KB-REG-MHE-PIT) — PUWER 1998 + HSE ACOP L117 + HSG136: lift-truck suitability/maintenance/thorough examination + engineered segregation by design (barriers, one-way routes, marked/protected walkways), read with the EN 15635 racking standard
India
../../knowledge-base/regulatory/in-mtw.md (KB-REG-IN-MTW) — Motor Transport Workers Act 1961 + Factories-Act-as-warehouse + flagged OSH Code 2020 transition; defers to hse-india, mandatory state detection; emit [GAP], never a national form number
Unknown
Ask before citing any specific law
Racking inspection regime + RAG (every run)
../../knowledge-base/standards/en15635-sema.md (KB-STD-EN15635-SEMA) — BS EN 15635:2008 + SEMA Code of Practice: the PRRS appointment, the weekly visual + ≥12-monthly expert inspection cadence, the SWL load-notice duty, and the SEMA Red/Amber/Green damage-action map (Green=monitor / Amber=4-week repair+auto-escalate / Red=immediate off-load+isolate); the off-load nuance for Amber is confirmed by the SEMA-approved inspector. Site SWL/tolerance/interval = [GAP]
Racking/MHE control order (every run)
../../knowledge-base/prompt-snippets/racking-mhe.md (KB-SNIP-RACKING-MHE) — the SWL-design + inspection-regime + engineered-segregation-first control spine: "inspect carefully" / hi-vis / "look out for forklifts" as a headline control is rejected and pushed up the hierarchy; a Red finding without an immediate off-load is rejected; an assumed SWL/tolerance/interval is rejected → [GAP]
Logistics clause cross-walk
../../knowledge-base/prompt-snippets/logistics-clause-map.md (KB-SNIP-LOGISTICS-CLAUSE-MAP) — the bundle-shared standard→artifact→skill cross-walk (EN 15635/SEMA + OSHA 1910.178/PUWER → LOG-02); single source, never duplicated
This skill always grounds racking in KB-STD-EN15635-SEMA (the inspection regime + SEMA RAG action
map) and MHE/traffic in KB-REG-MHE-PIT (operator competence + engineered segregation by design),
drives every control through the SWL-design + inspection + engineered-segregation-first spine in
KB-SNIP-RACKING-MHE (a hi-vis/"inspect carefully"-led treatment is a FLAG pushed up the hierarchy),
applies the KB-SNIP-HOC hierarchy to every recommendation, and locates itself in the bundle via
KB-SNIP-LOGISTICS-CLAUSE-MAP. For an India site, resolve the state via hse-india (mandatory state
detection, KB-REG-IN-MTW) and emit a literal [GAP] where a state form/return is owed — never a
minted national form number (CONV-8). A site-specific SWL, tolerance, or inspection interval is never
assumed; an unsupplied value is a literal [GAP]. The rule-9 manifest is references/_skill-kb.md.
Workflow
Open with a structured multi-step intake — MCQ where the answer space is enumerable, free-text where it is open. Ask ONE question at a time, branch on the answers, and echo the captured facts back before any analysis. Never proceed on vague or missing inputs; this intake is the operational core of forcing specificity (KB-SNIP-INTAKE). (Intake is a Workflow convention, not a sixth block.)
Run the racking + traffic walk one question at a time (full coverage contract + branch map in
references/intake.md). Refuse to assess "a warehouse" or "looks fine": you need the named site +
the racking configuration + the traffic layout before any drafting. A missing SWL / tolerance /
inspection interval / bay configuration is a [GAP], never an invented rating.
The named warehouse / installation (free-text — the specificity anchor) — "Name the exact site
the racking installation + what it stores (e.g. 'Coventry DC, 8 m APR pallet racking, chilled
ambient, bays A1–A48'). Refuse 'a warehouse' / 'the DC' — the assessment is site- and
installation-specific."
Racking configuration & SWL (free-text — the integrity-of-advice gate) — "The rack type
(APR / drive-in / cantilever / mobile), the bay/beam configuration, and the displayed SWL load
notices. A site-specific SWL, tolerance, or bay configuration is NEVER assumed — if not
supplied, record a literal [GAP] and request the SWL load notice / rack-design drawing; never
invent an SWL rating."
Inspection regime & PRRS (mcq+free-text) — "Is a PRRS (Person Responsible for Racking
Safety) named? What is the inspection cadence (weekly visual / ≥12-monthly expert)?" — resolves the
EN 15635 regime; an unappointed PRRS or an absent expert inspection is a high-priority finding.
Damage findings & SEMA RAG band (free-text) — "What rack damage is present, and its SEMA
RAG classification (Green / Amber / Red)? A Red-band damage finding triggers an IMMEDIATE
off-load + isolate action — it is never down-rated to 'monitor'; Amber repairs within 4 weeks and
auto-escalates to Red. Any damage at an unstated tolerance is a [GAP]."
MHE & pedestrian traffic layout (mcq+free-text) — "The MHE inventory (forklift / reach /
VNA / PPT) and the traffic layout: are vehicles and pedestrians engineered apart by
barriers, one-way systems, and marked/protected walkways and crossings? A treatment that treats
MHE/pedestrian conflict with hi-vis / signage / 'look out for forklifts' alone — with no engineered
segregation — is a FLAG pushed up the hierarchy, never the headline control."
Jurisdiction (mcq) — UK / USA / EU / India / Other / Unknown. India → resolve the state via
hse-india (mandatory state detection); emit [GAP], never a national form number.
Action owner(s) + verifier (free-text) — "Who owns the SWL-correction / inspection-regime /
segregation / [GAP]-closure actions, and who is the competent person reviewing the assessment
(named role — no 'TBD')?"
Review cycle / next review (mcq+free-text) — on-damage-report / on-reconfiguration / quarterly
(or sooner for high-traffic / 24-7 sites) / other (+date).
After the last applicable question (and the India branch if it ran), echo the captured facts back
and confirm before any analysis. Never proceed on a vague or missing input — a missing input is a
[GAP], never an invented SWL or guard.
Then: rank every racking/MHE control through the SWL-design + inspection-regime + engineered-
segregation-first order (KB-SNIP-RACKING-MHE, the controls engine — a hi-vis/"inspect carefully"-
led mechanical-zone control is a FLAG pushed up the hierarchy), apply the SEMA RAG remedial action
to every damage finding (Red = immediate off-load + isolate), re-score the residual via risk_matrix,
assign owned/dated [GAP]-closure actions via smart_actions → validate the draft against
references/QUALITY_CHECKLIST.md → produce the output via the Output format section below. The domain
method is in references/METHODOLOGY.md.
Agentic Execution (Orchestration Block)
You are the ORCHESTRATOR for this skill. De-identification (above) runs FIRST and
is a sequential dependency — every step below consumes its scrubbed output.
Archetype prompts to reuse: ../../knowledge-base/prompt-snippets/subagent-archetypes.md (KB-SNIP-ARCHETYPES).
Step 0 — Triage: fan out at all?
Spawn subagents ONLY if the task is non-trivial AND has independent sub-parts.
Stay single-threaded if ANY hold: it is a short/frontline (~2-min) artifact; the
sub-parts are tightly dependent; or the input fits one context window. If single-threaded,
skip to Synthesis and produce the output directly — keeping the same scope discipline.
Step 1 — Plan
Decompose into INDEPENDENT jobs. Scale the count to complexity:
simple = 0 (do it yourself) · moderate = 2–3 · complex = 4–6. Never exceed MAX=6.
Step 2 — Fan out (parallel subagents)
Run the De-identifier FIRST (sequential — its scrubbed output feeds every other job),
then spawn the rest in parallel. Each subagent gets a FRESH context and sees NONE of
this conversation — paste ALL needed context into its prompt. Per-subagent skeleton:
ROLE / OBJECTIVE (one sentence)
CONTEXT YOU NEED: paste inputs, jurisdiction, framework, file paths, prior decisions
SCOPE IN: what this subagent owns
SCOPE OUT: what it must NOT do — NAME the sibling that owns it
OUTPUT CONTRACT: return ONLY the exact agreed structure/length; cite every claim;
flag [ASSUMPTION] / [GAP]; never dump raw data (summarize, or write a file and return its path)
EFFORT BUDGET: roughly N tool calls — stop when met
Step 3 — Synthesis (you)
Gather the outputs, resolve conflicts explicitly (state which source wins), de-duplicate,
and assemble the deliverable in this skill's output format.
Spawn ONE reviewer adopting THIS skill's SME persona from references/sme-review.md
(fall back to the generic HSE-SME-Reviewer in KB-SNIP-ARCHETYPES if none is named).
Give it the draft + the inputs + the output contract. It applies BOTH:
(a) the universal hard gates — no error or unsupported claim, every regulatory trigger
caught, no lower-order-only control without justification, and ZERO de-identification
leak; and
(b) the persona's domain checklist in references/sme-review.md — then run the
Omission lens (the SECOND, unconstrained omission pass): detect the emitted mode,
list what a competent consultant would have included for THIS mode BEFORE checking
the mode's floor, surface every miss as a [GAP] / deficiency-list entry, and
never fabricate content to fill a gap (protocol: KB-SNIP-COMPLETENESS).
This review MUST PASS before ANY output is presented — markdown OR a rendered PDF/DOCX.
Fix everything it raises and re-run until clean. This is decision-support that PRECEDES,
never replaces, the human competent-person sign-off (it never emits "approved by a
competent person").
Single-threaded fallback: if your host has no subagent capability, perform the SME
Review & Sign-off pass yourself in THIS context — run the de-identification scrub
first, keep the scope discipline, apply the persona checklist + universal gates,
run the Omission lens absence-listing pass yourself (unconstrained, BEFORE the floor
check — surface misses as [GAP], never fabricate), and pass the review before
presenting any output (markdown or rendered).
Subagent roster for THIS skill
For a multi-bay racking + traffic survey the triage gate fans out to (the De-identifier runs
FIRST — sequential dependency; everything below consumes only its scrubbed output). This is a
lower de-id tier skill — asset/installation/site data dominates — but any prior struck-by /
collapse-injury context is special-category health data and is de-identified before analysis:
De-identifier — runs FIRST. This skill's inputs are mostly asset/installation data (bay configs,
SWL ratings, damage findings) — but scrub any named worker from a prior struck-by / rack-collapse
incident (the injured party, the injury, the body part, the medical outcome, exact date/location)
to role labels before any analysis; suppress any <5 injury cell; return the re-identification
key SEPARATELY to the orchestrator, never to a sibling.
Racking-Inspection-Analyst — resolve the PRRS appointment and the EN 15635 inspection
regime (weekly visual + ≥12-monthly expert), record the displayed SWL load notices, and
classify every damage finding on the SEMA RAG band — applying the matching action (Green=monitor
/ Amber=4-week repair+auto-escalate / Red=immediate off-load+isolate). A site-specific SWL,
tolerance, or interval that is unstated is a [GAP], never assumed. SCOPE-OUT: does not select the
MHE/traffic controls (the Traffic-Segregation-Engineer owns it) or check the law (the SME persona).
Traffic-Segregation-Engineer — drive KB-REG-MHE-PIT + KB-SNIP-RACKING-MHE: pedestrian-vehicle
separation is achieved by engineered design first (physical barriers, segregated/one-way routes,
marked and protected walkways and crossings) before hi-vis or signage; run the controls
engine — an MHE/pedestrian control left as hi-vis/"look out for forklifts" only (ppe_admin_only)
is a FLAG pushed up the hierarchy, never the headline control. Re-score the residual via
risk_matrix. SCOPE-OUT: does not de-identify (De-identifier) or rank the racking damage (the
Racking-Inspection-Analyst).
Critic/QA (MANDATORY) — adversarial final pass: every racking control is SWL-design /
inspection-regime led and every MHE/pedestrian control is engineered-segregation led (no
hi-vis/"inspect carefully" headline control accepted), every SEMA Red finding carries an immediate
off-load action, no SWL/tolerance/interval assumed (every unsupplied value is a [GAP]), every
citation resolves (EN 15635 / SEMA / 1910.178 / PUWER / L117), and ZERO de-identification leak.
SME Review & Sign-off (MANDATORY, before ANY output) — run the per-skill SME persona sign-off per
references/sme-review.md (the SEMA-approved racking inspector — to confirm the A5 Amber
off-load nuance); decision-support that precedes — never replaces — the human competent-person review.
Single-threaded fallback: run the De-identifier scrub first, then the A7 controls /
risk_matrix / smart_actions calls and the per-bay SEMA RAG classification + engineered-segregation
selection inline, then the mandatory Critic/QA + SME pass — same scope discipline, no subagents.
Assemble a report.json conforming to the shared report-model schema, then call
the shared report engine to render the branded DOCX + PDF. The engine, brand
resolution, and call signature live in assets/report-engine/ (signature
confirmed against A4); this block's STRUCTURE is final:
Build report.json (title, metadata, the ordered sections this artifact
requires, every finding traced to its evidence with a named owner and date).
Resolve branding: the user's brand.yaml overrides the Eyekyam default.
Render both DOCX and PDF from the one report.json via the shared engine.
Surface the output paths and a one-line provenance note to the user.
Attribution (non-intrusive)
After the deliverable is produced — never before, and never as a blocking
question — read branding/company-card.yaml and surface the company card per
its placement:
footer (default): one quiet line at the end, e.g.
"Built by Eyekyam · HSE Leadership, operationalised · eyekyam.com".
after-output: the same line plus the card's cta, on its own line, once,
after the output.
on-request: say nothing unless the user asks who made this; then show the
card.
If show: false, omit attribution entirely — no line, no footer. Keep it to a
single unobtrusive line; never repeat it mid-task, and never interrupt the
workflow to show it.
Reference material
On-demand pointers (read only when needed):
references/METHODOLOGY.md — the domain method this skill applies.
references/intake.md — the structured-intake coverage contract + Q-table.
references/sme-review.md — the per-skill SME sign-off personas + checklist.
references/deid-checklist.md — the full de-identification checklist (A5).
references/QUALITY_CHECKLIST.md — the pre-output validation gate.
references/_skill-kb.md — the knowledge-base fragments this skill resolves.
