Produces a consultant-grade, transmission-based infection prevention and control (IPC) plan for a named healthcare service or unit, driven by the engineering-and-administrative-controls-before-PPE hierarchy. Use this skill when a user asks to build or review an infection control plan, write an isolation / transmission-based precautions plan, select contact / droplet / airborne precautions for a ward or care home, decide device reprocessing by Spaulding classification, or plan de-identified outbreak / HAI surveillance. It forces Standard Precautions for every patient and the route-correct precautions, leads with engineering controls (ventilation, negative-pressure isolation rooms) and administrative controls (cohorting, screening) BEFORE PPE, applies the Spaulding reprocessing decision, and refuses a PPE-only plan (an airborne agent on a respirator alone, no isolation room). Grounded in the CDC isolation guideline + WHO IPC core components + Spaulding. Decision-support only; a competent person must review.
Produces a consultant-grade, transmission-based infection prevention and control (IPC) plan for a named healthcare service or unit, driven by the engineering-and-administrative-controls-before-PPE hierarchy. Use this skill when a user asks to build or review an infection control plan, write an isolation / transmission-based precautions plan, select contact / droplet / airborne precautions for a ward or care home, decide device reprocessing by Spaulding classification, or plan de-identified outbreak / HAI surveillance. It forces Standard Precautions for every patient and the route-correct precautions, leads with engineering controls (ventilation, negative-pressure isolation rooms) and administrative controls (cohorting, screening) BEFORE PPE, applies the Spaulding reprocessing decision, and refuses a PPE-only plan (an airborne agent on a respirator alone, no isolation room). Grounded in the CDC isolation guideline + WHO IPC core components + Spaulding. Decision-support only; a competent person must review.
A consultant-grade, engineering-and-administrative-controls-before-PPE transmission-based
infection prevention and control (IPC) plan for a named healthcare service or unit — a ward,
clinic, community/home-care service, care home, dental practice, ambulance service, or clinical
laboratory — never a generic "a hospital". Its entire reason to exist is that : every plan applies , identifies the of the agent(s) in scope,
and layers the correct — with
(ventilation, negative-pressure / airborne-infection isolation rooms, single rooms, safer devices)
and (cohorting, screening, signage, restricted entry) applied
PPE, which is the . A bare "give everyone an N95 / wear a mask" for an
airborne agent with is — a respirator is
the residual control, never the primary one where the route demands an engineering control.
## Output format
IPC leads with
ventilation, isolation, and administrative controls, not with PPE
Standard
Precautions to every patient
transmission route(s)
Transmission-Based precautions by route
engineering controls
administrative controls
before
residual barrier, last
no isolation room or ventilation provision
refused
It forces the single lever that separates a defensible artifact from copy-paste paperwork:
the route-correct precautions plus the full hierarchy of controls (engineering/administrative
before PPE), with no case or cluster ever identifying a patient. Healthcare infection data is
special-category health data (PHI) — a patient's infection / colonisation status and any
outbreak / cluster detail are de-identified to role / aggregate level before drafting, the
surveillance in the circulated artifact is de-identified and aggregated (never line-level), and
a case category with fewer than 5 individuals is suppressed (small-cell back-calculation
guarded — a 3-case cluster on a named ward re-identifies patients). The skill follows the
documented-procedure PHI model: it emits no re-identification key file (the key is an
instruction to the competent person, held separately) and adds no new report field — the
de-identified banner is the existing report meta.deid_notice header. Grounded in the CDC
Guideline for Isolation Precautions (Standard + Transmission-Based: contact / droplet / airborne),
the WHO Core Components of IPC Programmes, and the Spaulding classification (critical →
sterilization / semi-critical → high-level disinfection / non-critical → low-level), with the UK
Health and Social Care Act 2008 Hygiene Code, the OSHA bloodborne-pathogens confidentiality
discipline, and India Bio-Medical Waste Management Rules 2016 via hse-india. Decision-support
only; a competent person (infection-prevention & control professional) must review the output.
When to use this skill
Use this skill when the user needs an infection control / isolation plan for a concrete healthcare
service — for example "build an IPC plan for the respiratory ward", "write the transmission-based
precautions for a suspected TB admission", "select contact / droplet / airborne precautions for the
care home", "structure our IPC programme against the WHO core components", "decide how to reprocess
the endoscopes by Spaulding", or "set up de-identified HAI surveillance". It is not for a generic
"how do I stop infections?" answer: the Workflow intake below forces the named service, the agent(s)
and their transmission route(s), and the engineering/administrative controls before PPE, refuses
a vague "a hospital" request, and refuses a PPE-only treatment (an airborne agent managed by a
respirator alone with no isolation room or ventilation provision).
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.
This skill is one of the catalog's highest-PHI artifacts. The reinforced healthcare PHI
extension — the patient infection/colonisation-status rule, the outbreak / cluster
confidentiality rule, the <5 small-cell suppression with secondary back-calculation guard
(a 3-case cluster on a named ward re-identifies patients), and the
re-identification-key-separation instruction (documented-procedure model — no key file, no new
report field; the banner is the existing meta.deid_notice header) — lives in
references/deid-checklist.md and the Workflow de-id step below (it is NOT in the byte-identical
block above).
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
IPC framework spine (every run)
../../knowledge-base/standards/ipc-cdc-who.md (KB-STD-IPC-CDC-WHO) — the CDC isolation precautions + WHO IPC core components + Spaulding topic→artifact structure map: Standard + Transmission-Based precautions (contact/droplet/airborne), the WHO programme core components, and the critical/semi-critical/non-critical reprocessing decision (cite the topics, never paste the guideline)
Per-route precaution selection (every run)
../../knowledge-base/prompt-snippets/ipc-precautions.md (KB-SNIP-IPC-PRECAUTIONS) — the Standard-always → identify route → layer Transmission-Based precautions → engineering/administrative before PPE gate: a plan that jumps to PPE without the room/ventilation control the route requires, or an airborne agent with no AIIR/respirator, is rejected; cluster reporting is small-cell-suppressed
Healthcare clause cross-walk (every run)
../../knowledge-base/prompt-snippets/healthcare-clause-map.md (KB-SNIP-HEALTHCARE-CLAUSE-MAP) — the bundle-shared ISO 45001 6.1.2 + clinical PPE-last cross-walk that keeps the five hse-healthcare skills consistent
USA / bloodborne-exposure confidentiality
../../knowledge-base/regulatory/osha-bbp.md (KB-REG-OSHA-BBP) — OSHA 29 CFR 1910.1030 confidentiality discipline for exposure / health data ((f) confidential post-exposure follow-up) reused for IPC PHI handling (cite the paragraph topics, never paste the rule)
India
../../knowledge-base/regulatory/in-bmw2016.md (KB-REG-IN-BMW2016) — India Bio-Medical Waste Management Rules 2016 segregation (yellow/red/blue/white-translucent stream); defers to hse-india, mandatory state detection; emit [GAP], never a national form number
Unknown
Ask before citing any specific law
This skill always grounds in KB-STD-ISO45001 (6.1.2 / 8.1.2) and the IPC framework spineKB-STD-IPC-CDC-WHO (CDC isolation precautions + WHO core components + Spaulding), selects precautions
by route through KB-SNIP-IPC-PRECAUTIONS (Standard always → engineering/administrative controls
before PPE), aligns with the other hse-healthcare skills through KB-SNIP-HEALTHCARE-CLAUSE-MAP,
applies KB-SNIP-HOC to every control, and reuses the KB-SNIP-ARCHETYPES subagent roster. The
precaution selection is a structured route-driven method over the named unit + agents + the cited
CDC/WHO/Spaulding frameworks — not a calculation (no new engine). For US exposure / health-data
confidentiality ground in KB-REG-OSHA-BBP; for India, resolve the state via hse-india
(mandatory state detection) per the BMW Rules 2016, and emit a literal [GAP] where a state
return is owed — never a minted national form number. 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.)
De-identify FIRST (the highest-PHI step — before any drafting). Run the deid block +
references/deid-checklist.md BEFORE the intake echo-back drives any analysis. Healthcare infection
data is special-category health data: scrub any patient's infection / colonisation status (never
attributed to a named person in a circulated plan), any outbreak / cluster detail (reported
de-identified and aggregated), and any patient/staff identifier anywhere in the inputs. Apply <5
small-cell suppression (with secondary suppression) to every case / cluster category — a 3-case
cluster on a named ward re-identifies patients (the small-cell back-calculation). Follow the
documented-procedure PHI model: the re-identification key is an instruction to the competent
person (held separately, access-controlled) — the skill emits no key file and adds no new
report field (the de-identified banner is the existing report meta.deid_notice header).
Run the IPC intake one question at a time (full coverage contract + branch map in
references/intake.md). Refuse to plan for "a hospital": you need the named service + the agent(s)
and their transmission route(s) before any precaution plan. Refuse a PPE-only treatment — an airborne
agent managed by a respirator alone with no airborne-infection isolation room or ventilation
provision is rejected and pushed up the hierarchy.
The named service & unit (free-text — the specificity anchor) — "Name the exact unit / service
(ward, clinic, care home, dental surgery, ambulance, lab). Refuse 'a hospital' / 'the ward' —
the plan is service-specific."
The agent(s) and their transmission route(s) (mcq+free-text — the routing anchor) — contact /
droplet / airborne / a combination (and the suspected/confirmed agent). The route drives the
precautions; an unknown route is a [GAP], resolved before precautions are selected.
Engineering controls available (free-text — asked BEFORE PPE) — ventilation, single rooms,
negative-pressure / airborne-infection isolation room (AIIR), safer devices. An airborne agent
with no AIIR or ventilation provision controlled by a respirator alone is FLAGGED and pushed up the
hierarchy.
Device reprocessing (Spaulding) (free-text) — which devices contact sterile tissue/bloodstream
(critical → sterilization), mucous membranes / non-intact skin (semi-critical → high-level
disinfection), or intact skin (non-critical → low-level). High-level disinfection of a critical
device is a Spaulding mis-application and fails.
Surveillance (free-text) — HAI / outbreak / cluster monitoring. Reported de-identified and
aggregated; every <5 case category suppressed (with secondary back-calc guard). Refuse to report
an outbreak on a named ward in a way that re-identifies a patient.
Jurisdiction (mcq) — USA / UK / EU / India / Other / Unknown. India → resolve the state via
hse-india (mandatory state detection); BMW Rules 2016 segregation; emit [GAP], never a national
form number.
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 agent, route, or count.
Then: apply Standard Precautions to every patient, identify the transmission route(s), and
layer the Transmission-Based precautions by route (KB-SNIP-IPC-PRECAUTIONS); rank the controls
up the hierarchy via the controls engine (engineering — ventilation/AIIR/single rooms — →
administrative — cohorting/screening/signage — → PPE last; a PPE-only / respirator-alone treatment
where the route demands an engineering control is a FLAG pushed up the hierarchy); record the
Spaulding reprocessing decision per device class; structure the WHO IPC programme (core
components); build the de-identified / aggregated surveillance with <5 small-cell suppression;
frame the residual transmission risk via risk_matrix; make every action a SMART action 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-route IPC plan the triage gate fans out (moderate 2–3; the De-identifier runs FIRST —
sequential dependency; everything below consumes only its scrubbed output):
De-identifier — runs FIRST. Scrub every patient/staff identifier — esp. any patient's
infection / colonisation status and any outbreak / cluster detail — to role / aggregate level
before any analysis; apply <5 small-cell suppression (with secondary suppression) to every case /
cluster category (a 3-case cluster on a named ward re-identifies patients); the re-identification
key is an instruction returned SEPARATELY to the orchestrator, never to a sibling, and never as a
key file.
Precautions-&-Controls-Analyst — applies Standard Precautions to every patient, identifies
the transmission route(s), and layers the Transmission-Based precautions by route via the
controls engine: lead with engineering controls (ventilation, AIIR / negative pressure, single
rooms) and administrative controls (cohorting, screening, signage), with PPE as the residual
barrier last. A treatment that leads with PPE — an airborne agent managed by a respirator alone
with no isolation room or ventilation — is a FLAG pushed up the hierarchy, never the headline
control. SCOPE-OUT: does not author the surveillance / programme structure (the
Surveillance-&-Programme-Author) or de-identify (the De-identifier).
Surveillance-&-Programme-Author — structures the WHO IPC programme (core components) and the
de-identified / aggregated surveillance (HAI / outbreak / cluster — never line-level, every <5
category suppressed with the secondary back-calc guard), and records the Spaulding reprocessing
decision per device class. No outbreak is reported on a named ward in a way that re-identifies a
patient. SCOPE-OUT: does not select the precautions/controls (the Precautions-&-Controls-Analyst)
or de-identify (the De-identifier).
Critic/QA (MANDATORY) — adversarial final pass: Standard Precautions are applied to every
patient, the route is identified and the correct Transmission-Based precautions layered, the
engineering and administrative controls are recorded BEFORE PPE (a respirator-alone airborne plan
is flagged), the Spaulding reprocessing decision is correct (no high-level disinfection of a critical
device), the surveillance is de-identified / aggregated, no <5 case cell is published (with the
secondary back-calc guard), 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; decision-support that precedes —
never replaces — the human competent-person review.
Single-threaded fallback: run the De-identifier scrub first, then Standard + the route-correct
Transmission-Based precautions, the A7 controls / risk_matrix / smart_actions calls inline, the
Spaulding decision + the WHO programme structure + the de-identified surveillance, 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) + the healthcare PHI extension.
references/QUALITY_CHECKLIST.md — the pre-output validation gate.
references/_skill-kb.md — the knowledge-base fragments this skill resolves.
