contractor-prequalification

name	contractor-prequalification
description	Produces a risk-tiered contractor prequalification questionnaire (PQQ) and an evidence-based scoring/evaluation of a contractor's HSE capability for a named scope of work. Use this skill whenever a user asks to prequalify, vet, or approve a contractor or subcontractor, build a PQQ or contractor-approval questionnaire, score a contractor's safety credentials, or set up contractor HSE management. It tiers the rigour to the work's risk, asks for verifiable evidence (not claims), scores against defined criteria, and produces an approve/conditional/reject recommendation with named conditions and review dates — emitted as a branded report. Decision-support only; a competent person must review the output.
license	Apache-2.0
metadata	{"author":"eyekyam","version":"1.0","category":"contractor","tier":2,"audience":["M","C"],"industry":["All"],"jurisdiction":["All"],"status":"stable","plugin":"hse-operations","hse_reviewed_by":"","hse_reviewed_date":""}

Contractor Prequalification

A consultant-grade HSE skill that produces a risk-tiered contractor prequalification questionnaire (PQQ) and an evidence-based scorecard of a contractor's HSE capability for a named scope of work, grounded in ISO 45001 clause 8.1.4 (procurement / contractors / outsourcing). It forces the single lever that separates a defensible appointment from a rubber-stamp: it tiers the rigour of questioning to the work's risk and scores only against verifiable evidence — never a self-asserted claim. Missing or unverifiable evidence becomes a [GAP] that blocks an unconditional pass; high-hazard work cannot pass on PPE/competence claims alone. The output is an approve / conditional / reject recommendation with named conditions, owners, and a review date — emitted as a branded report.

The risk-tier → PQQ-depth → pass-threshold map (KB-DATA-CONTRACTOR-TIERS) and the evidence-anchored question bank (KB-SNIP-PQQ-BANK) are cited rubrics applied as structured prose — there is no contractor_score engine; the discipline is that every score is tied to a verified evidence item.

When to use this skill

Use this skill when the user needs to prequalify, vet, or approve a contractor or subcontractor for a concrete scope of work on a named site — for example "prequalify acme scaffolding for the work-at-height package on the Plant 3 turnaround", "build a PQQ for our mechanical-install subcontractor", "score this contractor's safety credentials", or "set up contractor HSE management for the demolition phase". Trigger phrases: contractor prequalification, PQQ, contractor approval, vet a subcontractor, contractor HSE management, SSIP, approve/reject a contractor, contractor scorecard. It tiers the depth of questioning to the work's risk (T1 low/admin · T2 medium · T3 high-hazard), asks for verifiable evidence not claims, and produces an approve/conditional/reject recommendation. If the request is vague ("approve this contractor"), the Workflow intake below refuses to score until the scope+risk tier and the core evidence set 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.

Jurisdiction / scope	Read
India	../../knowledge-base/regulatory/in-factories-act.md (+ in-state-forms.md for the user's state)
UK	../../knowledge-base/regulatory/uk-hswa.md (HSWA s.3/s.4 + CDM 2015 reg. 8 where the work is construction)
USA	../../knowledge-base/regulatory/us-osha.md (OSHA multi-employer worksite policy CPL 02-00-124 — controlling-employer duty)
EU	../../knowledge-base/regulatory/eu-osh.md
Unknown	Ask before citing any specific law

This skill always grounds in KB-STD-ISO45001 clause 8.1.4 (procurement / contractors / outsourcing) and applies KB-SNIP-HOC to every control it evaluates in the contractor's proposed safe system of work. It sets PQQ depth and the pass threshold from the risk-tier map KB-DATA-CONTRACTOR-TIERS (T1 low / T2 medium / T3 high-hazard — tier set by the highest-risk activity in the scope), and draws the evidence-anchored questions from KB-SNIP-PQQ-BANK (each question tied to its required evidence). It locates itself in the bundle via the operations clause cross-walk KB-SNIP-OPS-CLAUSE-MAP (8.1.4 → contractor-prequalification). For an India site it resolves the state via KB-REG-IN-STATEFORMS (mandatory state detection — never a national form number). The rule-9 manifest is references/_skill-kb.md. Quote each fragment's source+year.

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 (scope & risk tier Q1 · contractor & work package Q2 · evidence available Q3 · jurisdiction Q4 · decision type Q5), the risk-tier branch (Q1 = high-hazard → the T3 high-hazard PQQ module + the higher pass threshold), the mandatory India→state branch (Q4 = India → Q4a), 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 score until the scope+risk tier and at least the core evidence set (insurance + accident/enforcement history + relevant competence) are captured. Q1 (scope & risk) is the specificity anchor — it sets the tier, the PQQ depth, and the pass threshold; Q3 missing evidence becomes a [GAP] (never invent, never score a self-asserted claim). A [GAP] blocks an unconditional pass.

The prequalification method (ISO 45001 8.1.4 — risk-tiered, evidence-anchored)

Full method in references/METHODOLOGY.md. Steps:

1. De-identify the inputs FIRST — before any scoring (the deid block above + the De-identifier-runs-first orchestration rule). A contractor's accident / enforcement history can name injured persons (prosecutions, RIDDOR/OSHA records); scrub names, exact dates, and locations to role labels and apply small-cell suppression to a small contractor's injury counts before the evidence reaches the scorer. Everything downstream consumes only the scrubbed text.
2. Assign the risk tier — from Q1, band the scope into T1 / T2 / T3 per KB-DATA-CONTRACTOR-TIERS. The tier is set by the highest-risk activity in the scope, not the average — a single hot-work / confined-space / work-at-height / lifting / energised-electrical / demolition task lifts the whole package to T3. The tier sets the PQQ depth and the pass threshold.
3. Build the tiered PQQ — select the question set for the tier from KB-SNIP-PQQ-BANK (Core for T1+ → Task-specific for T2+ → the high-hazard module for T3). Each question names the verifiable evidence that answers it — a signed policy, a current insurance certificate, the RAMS, an in-date accreditation certificate (SSIP / ISO 45001), the activity-specific competence certification, the de-identified accident-rate data.
4. Build the evidence register & verify — for every PQQ item, record the claim, the evidence supplied, and a verification status (verified / unverified / not-supplied). A self-asserted claim with no supporting evidence is [GAP] — never scored as met (the core integrity gate). Evaluate the contractor's proposed controls against the hierarchy via controls.rank_controls / controls.validate_treatment + KB-SNIP-HOC: high-hazard (T3) work cannot pass on PPE/competence claims alone — higher-order controls must be evidenced.
5. Score against the threshold — score each criterion only on its verified evidence, then apply the tier's pass threshold (KB-DATA-CONTRACTOR-TIERS). This is cited-rubric structured prose, not a deterministic engine (§8 anti-sprawl — there is no contractor_score engine). Any score must trace to a named evidence item; an unverified or [GAP] criterion cannot clear the threshold.
6. Recommend approve / conditional / reject — approve only when every threshold criterion is met on evidence; when gaps are closeable — each condition is a (: named owner + ISO due date + measure, tied to the gap it closes) ; when a threshold criterion fails irrecoverably (e.g. unresolved serious enforcement, no high-hazard competence for T3 work). Never "approve and proceed" over an open .

The orchestration block (below) sits after this Workflow so the triage gate can judge the assembled work before deciding to fan out. A single low-risk (T1) contractor runs single-threaded; a complex multi-package or high-hazard evaluation fans out (the De-identifier runs first as a sequential gate). The de-identification and evidence-verification steps are mandatory in every case, single-threaded or fanned-out.

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 moderate roster (A6 "moderate = 2–3") for a complex multi-package or high-hazard contractor evaluation. The De-identifier is the sequential first gate (not a fan-out peer); there is no contractor_score subagent — the scoring is the cited weighted rubric (KB-DATA-CONTRACTOR-TIERS / KB-SNIP-PQQ-BANK) applied as structured prose by the Scorer-&-Recommender, with the hierarchy check via the A7 controls engine and the conditions via smart_actions. Archetypes: KB-SNIP-ARCHETYPES.

• De-identifier — runs FIRST (sequential gate, not a fan-out peer); scrub the contractor's accident/enforcement history — named injured persons, witnesses, exact dates/locations — to role labels, and suppress any injury cell < 5 on a small contractor, before any analysis. Returns the re-identification key SEPARATELY (to the orchestrator, never to a sibling). Every job below consumes only its scrubbed output.
• Evidence-Verifier — for each PQQ item, checks the contractor's claim against the supplied evidence and records the verification status; flags [GAP] for every unverified or not-supplied claim (a self-asserted accreditation with no certificate is a [GAP], never "met"). SCOPE-OUT: does not score or recommend (Scorer-&-Recommender owns it), does not draft the report.
• Scorer-&-Recommender — assigns the risk tier (KB-DATA-CONTRACTOR-TIERS), scores only on verified evidence against the tier's pass threshold, evaluates the proposed controls up the hierarchy via controls (T3 cannot pass on PPE/competence alone), and produces the approve / conditional / reject recommendation with conditions as dated smart_actions + a review date. SCOPE-OUT: does not re-verify evidence (Evidence-Verifier owns it) or scrub PII (De-identifier owns it).
• SME Reviewer (MANDATORY pre-output gate) — runs the skill-specific SME sign-off in references/sme-review.md (Procurement HSE Lead) before any output: the tier is set by the highest-risk activity, no score rests on a self-asserted claim, high-hazard work is not passed on PPE/competence alone, and every condition is owned + dated.
• Critic/QA (MANDATORY) — adversarial final pass: every score tied to a verified evidence item, every [GAP] blocks an unconditional pass, the hierarchy applied to the proposed controls, every condition owned + dated, ISO 45001 8.1.4 (and CDM reg. 8 / OSHA multi-employer where relevant) cited accurately, and zero PII leak (no named injured person, no < 5 cell, no re-id key in the output). PASS/FAIL.

A single low-risk (T1) contractor runs single-threaded — no subagents — but the de-identification scrub, the evidence-verification step, and the Critic/QA + SME pass 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.