Sector-tuned incident investigation for a mining event (a DGMS-reportable accident or dangerous occurrence) — de-id-first 4-fan-out, ICAM-led root-cause analysis reaching organisational factors, a DGMS reportability verdict (24-hour notice + Form J entry), and a hierarchy-of-controls CAPA plan with named owners and due dates. Use it to investigate a mine accident, run an RCA on a mining incident, or determine DGMS reportability. Decision-support only; a competent (DGMS-qualified) person must review the output. Statutory form ids beyond the verified DGMS anchors are recorded as [GAP], never invented.
Sector-tuned incident investigation for a mining event (a DGMS-reportable accident or dangerous occurrence) — de-id-first 4-fan-out, ICAM-led root-cause analysis reaching organisational factors, a DGMS reportability verdict (24-hour notice + Form J entry), and a hierarchy-of-controls CAPA plan with named owners and due dates. Use it to investigate a mine accident, run an RCA on a mining incident, or determine DGMS reportability. Decision-support only; a competent (DGMS-qualified) person must review the output. Statutory form ids beyond the verified DGMS anchors are recorded as [GAP], never invented.
A consultant-grade incident-management skill — sector-tuned incident investigation for a mining event (a DGMS-reportable accident or dangerous occurrence). It extends the B5 flagship pattern verbatim: a de-id-first 4-subagent fan-out (De-identifier -> Evidence/Timeline + Root-Cause Analyst + Regulatory-Reportability Checker + Corrective-Action Drafter -> mandatory Critic/QA), ICAM-led RCA reaching organisational factors, cause->evidence + CAPA->cause traceability, and a hierarchy-of-controls CAPA (controls / smart_actions). It overlays the — the register entry (), region-resolved (). Statutory form ids beyond the verified DGMS anchors are recorded , never invented. is context only. Decision-support only — a competent (DGMS-qualified) person reviews the output.
## Output format
DGMS reportability path
24-hour accident notice + Form J
KB-REG-IN-DGMS
KB-REG-IN-STATEFORMS
[GAP]
incident_rates
When to use this skill
Use this skill to investigate a mine accident, run an RCA on a mining incident, build a CAPA from a mining event, or determine DGMS reportability — for a specific mining incident. It is the core B5 investigation with a mining intake, ICAM as the default RCA method, and a DGMS 24h-notice + Form-J reportability path. De-identification runs FIRST. If the request is vague, the Workflow intake forces the incident facts first; any un-verified DGMS form id is recorded [GAP], never fabricated.
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.
Open with a structured multi-step intake — MCQ where the answer space is enumerable, free-text where it is open; one question at a time, branch on the answers, echo the captured facts before any analysis; refuse on a vague subject and never invent (KB-SNIP-INTAKE).
Step 0 — Structured intake (run this first, one question at a time)
De-identification runs FIRST (the orchestration block below) — injured/deceased-miner
identity, witness detail, exact pit/shaft locations and small (<5) fatality/injury cells are
scrubbed to role labels and suppressed before any analysis; the intake echo-back
shows personnel as role labels only. The full typed, branched Q-table — the mandatory DGMS
region/zone gate, the event-class branch (which gates DGMS reportability + de-id intensity),
the exposure counts, the de-identified facts, the evidence multi-select, the RCA-method
selection, and the echo-back / refuse-on-vague anchors — lives in references/intake.md.
Must-ask dimensions: the event class + datetime, the DGMS region/zone (mandatory
gate), the involved-persons counts (drives <5 suppression), and the RCA method
(ICAM default). Then (B5 pattern + DGMS overlay): reconstruct the timeline + evidence log;
run the RCA (rca, reaches_systemic); resolve DGMS reportability — the 24-hour
accident notice + Form J entry for a DGMS-reportable accident/dangerous occurrence
(KB-REG-IN-DGMS), region-resolved; cite only the verified DGMS anchors as values, any
other form id [GAP], never invented; draft the HoC-tagged CAPA (controls /
smart_actions), each tracing to a named cause with owner + due date. incident_rates is
context only. Validate against references/QUALITY_CHECKLIST.md, then produce the output. A
competent (DGMS-qualified) person reviews it.
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
De-identifier — runs FIRST (sequential dependency); scrub all PII/health detail —
injured-party, witnesses, diagnoses, exact dates/locations, small injury cells — into role
labels before any analysis. Returns the re-identification key SEPARATELY (to the orchestrator,
not to any sibling). Everything below consumes only its scrubbed output.
A · Evidence & Timeline Reconstructor — assemble the numbered, time-ordered event
sequence and the numbered evidence log (E-1…) from the scrubbed inputs; flag [GAP].
SCOPE-OUT: does not assign causes (B owns it) or decide reportability (C owns it).
B · Root-Cause Analyst — apply the chosen method (5-Whys / ICAM / SCAT / Fishbone /
Swiss-Cheese) via rca.py; every causal claim cites an evidence item (E-n); reach a
systemic/organisational factor (rca.validate reaches_systemic true for whichever method).
SCOPE-OUT: reportability (C owns it) and control selection (D owns it).
C · Regulatory Reportability Checker — for the resolved jurisdiction (India → resolved
STATE first), return verdict + clause/section + deadline + form (India state accident form
via KB-REG-IN-STATEFORMS / UK RIDDOR / US OSHA 29 CFR 1904). Conservative — flag [GAP] and
"ask a competent person" when unsure. SCOPE-OUT: does not draft the report or invent a form number.
D · Corrective-Action Drafter — hierarchy-of-controls-tagged CAPAs, each tracing to a
named cause (RC-n) with a named owner + ISO due date + measure; prefer higher-order controls,
justify any PPE/admin-only. SCOPE-OUT: does not score causes (B) or check law (C).
Critic/QA (MANDATORY) — adversarial read-only review: every cause evidence-backed, RCA
reaches a systemic factor, reportability cited conservatively to the matched KB row, every
CAPA traces to a cause with owner+date, no PPE/admin-only without justification, and ZERO
PII/health leak (no residual identifier, no <5 cell, no re-id key in the output).
Step 4 — SME review & sign-off (MANDATORY, before any output): run the skill-specific
SME persona sign-off in references/sme-review.md (the mining-incident lead investigator,
ICAM-trained) — model QA, decision-support, FLAGs non-blocking; it precedes and never
replaces the DGMS-qualified competent-person review / statutory inquiry.
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/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.
