business-continuity-plan

name	business-continuity-plan
description	Produces an ISO 22301 business continuity plan for a named organisation — business impact analysis, recovery objectives (RTO/RPO/MTPD), continuity strategies, and a plan with named recovery roles. Use this skill whenever a user asks to write a business continuity plan, BCP, BIA, disaster-recovery or resilience plan, or to set RTO/RPO recovery objectives. It runs a BIA to find time-critical activities and their dependencies, derives RTO/RPO within the MTPD (RTO < MTPD, never asserted alone), selects continuity strategies, and documents the plan with named owners, deputies, and review dates — emitted as a branded report. It complements (not duplicates) the emergency-response-plan: ERP handles the incident response; BCP handles continuity of critical activities. Decision-support only; a competent person must review the output.
license	Apache-2.0
metadata	{"author":"eyekyam","version":"1.0","category":"ms-admin","tier":3,"audience":["M","C","E"],"industry":["All"],"jurisdiction":["All"],"status":"stable","plugin":"hse-operations","hse_reviewed_by":"","hse_reviewed_date":""}

Business Continuity Plan

A consultant-grade HSE skill that produces a named-organisation ISO 22301 business continuity plan: it runs a Business Impact Analysis (BIA) to find the time-critical activities and their dependencies, derives recovery objectives — MTPD, then RTO strictly inside that MTPD (never an RTO asserted alone) plus an RPO — per critical activity, selects continuity strategies that cover every stated dependency, and documents the plan with named recovery roles, deputies, an exercise/test schedule, and review dates. It forces the single lever that separates a defensible artifact from copy-paste paperwork: specificity — every RTO/RPO/strategy traced to the BIA's evidence with an owner and a recovery role; preventive continuity controls are ranked by the hierarchy of controls (KB-SNIP-HOC), never a token treatment. Grounded in KB-STD-ISO22301 (clauses 8.2.2 BIA · 8.2.3 risk assessment · 8.3 strategies · 8.4 plans · 8.5 exercise programme). It complements, never duplicates, emergency-response-plan — ERP owns the immediate incident response (muster, evacuation, scenario procedures); BCP owns continuity of critical activities after the incident is controlled.

When to use this skill

Use this skill when the user needs a business continuity plan for a named organisation, site, or function — for example "write a BCP for our claims-processing operation", "run a BIA and set RTO/RPO for the order-fulfilment function", "build a disaster-recovery / resilience plan for the data centre", or "what's our MTPD for payroll and what continuity strategy covers a single-supplier dependency?". Trigger phrases: business continuity plan, BCP, BIA, business impact analysis, RTO, RPO, MTPD, recovery objectives, continuity strategy, disaster recovery, resilience plan.

BCP vs ERP (the D-07 boundary — disambiguate at intake Q1). If the request is about the immediate incident response — fire/evacuation, muster points, the call-out tree, scenario response procedures, emergency drills — that is emergency-response-plan, not this skill; route there. This skill owns the continuity leg: critical activities, MTPD/RTO/RPO, dependencies, recovery strategies, recovery roles. The two are adjacent and cross-referenced, never merged. If the request is vague ("write me a continuity plan"), the Workflow intake below refuses to proceed until the named critical activities and their dependencies are captured.

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.

1. 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.
2. 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.
3. 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.
4. 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.
5. 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.

This skill is grounded in a jurisdiction-independent management-system standard — ISO 22301 carries no national forms — so the table below is a fragment-resolution map, not a jurisdiction selector. Read only the fragment a given step needs:

What you need	Read (fragment)
ISO 22301 clause → artifact map (BIA · RTO · RPO · strategies · plans · exercise)	../../knowledge-base/standards/iso-22301.md (KB-STD-ISO22301)
The BIA → impact-over-time → MTPD → RTO/RPO derivation method	../../knowledge-base/prompt-snippets/bia-method.md (KB-SNIP-BIA-METHOD)
Recovery-objective banding + the RTO < MTPD margin discipline	../../knowledge-base/data-points/rto-rpo-guidance.md (KB-DATA-RTO-RPO-GUIDANCE)
Which hse-operations sibling owns an adjacent ISO 45001 operations clause	../../knowledge-base/prompt-snippets/ops-clause-map.md (KB-SNIP-OPS-CLAUSE-MAP)
The hierarchy of controls for any preventive continuity control	../../knowledge-base/prompt-snippets/hierarchy-of-controls.md (KB-SNIP-HOC)

This skill always grounds in KB-STD-ISO22301 and follows KB-SNIP-BIA-METHOD; it reuses the Phase-11-shipped KB-STD-ISO22301 and never re-authors it. The RTO < MTPD rule is a checklist constraint (KB-DATA-RTO-RPO-GUIDANCE, references/QUALITY_CHECKLIST.md), not a deterministic engine. For continuity of the incident-response leg, the BIA method itself carries a one-line cross-reference to emergency-response-plan — BCP complements, never duplicates, the ERP (KB-SNIP-OPS-CLAUSE-MAP, D-07). For a specific jurisdiction's statutory continuity duty (e.g. India Factories Act s.41B on-site emergency plan for MAH installations), ask the user before citing any specific law.

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.)

Step 0 — Structured intake (run this first, one question at a time)

The full typed, branched intake — the intake-coverage manifest, the question table, the BCP-vs-ERP disambiguation at Q1, the echo-back, and the refuse-on-vague anchors — lives in references/intake.md. Run it one question at a time, branch on the answers, echo the captured facts back before any analysis, and refuse to proceed on a vague scope. The five domain questions:

1. Q1 — Scope (free-text, the specificity anchor + BCP-vs-ERP gate). The named organisation / site / function this BCP covers. First disambiguate: if the user actually wants the immediate incident response (muster, evacuation, scenario procedures, the call-out tree, emergency drills), route to emergency-response-plan and stop — that is the ERP, not the BCP. This skill owns continuity of critical activities (critical activities / RTO / RPO / MTPD / recovery). Refuse a generic scope ("the whole company" with no named function).
2. Q2 — Critical activities + their outputs (free-text list). The time-critical activities and what each must keep producing. This drives the BIA; refuse to set any recovery objective until these are captured.
3. Q3 — Disruption scenarios (multi-select): loss of site · loss of IT · loss of key supplier · loss of key staff · utility failure — branch per selected scenario.
4. Q4 — Current recovery capability (MCQ): none / informal / documented-DR / tested-BCP.
5. Q5 — Objectives basis (free-text): any known MTPD/RTO/RPO, and the dependencies (people · IT/systems · suppliers · premises · equipment) per critical activity. RTO must be derived under a stated MTPD, never asserted alone (KB-DATA-RTO-RPO-GUIDANCE).

Refuse-on-vague anchor: no recovery objectives until the named critical activities and their dependencies are captured; an RTO with no stated MTPD basis, or RTO ≥ MTPD, is invalid.

Then: run the BIA method (references/METHODOLOGY.md / KB-SNIP-BIA-METHOD) → validate the draft against references/QUALITY_CHECKLIST.md → produce the output via the Output format section below.

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.

Step 4 — SME Review & Sign-off (MANDATORY — regulatory/safety output)

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

This is the STANDARD moderate roster (A6 "moderate = 2–4"). The De-identifier is the sequential first gate (not a fan-out peer); the fan-out jobs are BIA-Analyst → Strategy-&-Objectives → Plan-Assembler; SME Reviewer + Critic/QA are mandatory. Archetypes: KB-SNIP-ARCHETYPES. The exercise/test schedule and recovery actions are smart_actions calls (owners + ISO dates), not a fan-out job.

• De-identifier — runs FIRST (sequential gate, not a fan-out peer); scrub all PII/health detail into role labels before any analysis. Recovery roles legitimately name individuals, but a recovery-role holder's home address / medical note must never enter a shared copy — role-label it (every fan-out job below consumes scrubbed text).
• BIA-Analyst — from the scrubbed inputs, identify the critical activities + their outputs, map impact over time to derive each MTPD, and capture every dependency (people · IT · suppliers · premises · equipment); flag [GAP]. SCOPE-OUT: setting RTO/RPO or strategies (Strategy-&-Objectives owns it).
• Strategy-&-Objectives — derive RTO strictly under the stated MTPD (with margin) + RPO per activity (KB-DATA-RTO-RPO-GUIDANCE — an RTO with no MTPD basis, or RTO ≥ MTPD, is invalid), and select continuity strategies covering every stated dependency; preventive controls ranked via KB-SNIP-HOC. SCOPE-OUT: the BIA (BIA-Analyst), assembling the plan (Plan-Assembler).
• Plan-Assembler — assemble the plan: recovery roles each with a deputy, the exercise/test schedule (smart_actions, owners + ISO dates), the review cycle, and the one-line cross-reference to emergency-response-plan for the incident-response leg (BCP complements, does not duplicate, the ERP). SCOPE-OUT: deriving objectives (Strategy-&-Objectives).
• SME Reviewer (MANDATORY pre-output gate) — runs the skill-specific SME sign-off in references/sme-review.md (Business Continuity Manager (ISO 22301)): every RTO under a stated MTPD, every strategy covering the stated dependencies, every recovery role with a deputy, an exercise schedule present, and no recovery-role personal detail in a shared copy.
• Critic/QA (MANDATORY) — adversarial final pass: every recovery objective traced to the BIA, no RTO ≥ MTPD / no RTO without an MTPD basis, every dependency covered by a strategy, every citation traces to KB-STD-ISO22301, and ZERO PII/health leak. PASS/FAIL.

Simple single-function plans run single-threaded — no subagents — but the smart_actions call and the SME + Critic/QA passes are still made.

## Output format

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:

1. Build report.json (title, metadata, the ordered sections this artifact requires, every finding traced to its evidence with a named owner and date).
2. Resolve branding: the user's brand.yaml overrides the Eyekyam default.
3. Render both DOCX and PDF from the one report.json via the shared engine.
4. 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.