# ProcureAI

> Free, fine-tunable agentic-AI suite for procurement leaders. Sixteen Claude / GPT / Gemini-compatible skills across the four procurement pillars — sourcing, contracts, spend, suppliers — plus field-tested writing on what actually ships on procurement desks. By Martin Bacigal (procureai.tech).

ProcureAI publishes a free, openly-downloadable skill suite that turns a frontier LLM (Claude Opus 4.7, GPT-5.5, Gemini 3.1 Pro, etc.) into a working procurement co-pilot. Each skill ships with sensible defaults plus override files (`assets/playbook.yaml`, `references/internal-policies.md`) that customers replace to make the skill their own. No platform deal, no consulting upsell, no email gate on individual content pieces. Author: Martin Bacigal — Founder of ProcureAI, Global Category Manager — IT at Nouryon, ex-Henkel Global Center of Excellence; built ProcurementGPT (RAG + vector storage, +35% sourcing efficiency); trained 350+ procurement professionals across four continents.

## The 16-skill agentic procurement suite

- [Machine-readable catalog (JSON-LD)](https://procureai.tech/.well-known/skills.json): Full schema.org-extended catalog of every skill with trigger phrases, pillar, version, download URL, fine-tune surface, and model-routing defaults. Recommended entry point for autonomous agents.
- [Full-suite ZIP](https://procureai.tech/downloads/procureai-skills.zip): All 16 SKILL.md files plus assets and references in one bundle (~130 KB).
- [Suite README](https://procureai.tech/downloads/skills/README.md): Conventions, override-file structure, model-routing guidance.

### Sourcing & RFx (4 skills)
- [xRFP-drafter](https://procureai.tech/.well-known/skills.json#xrfp-drafter): Turn a one-paragraph category brief into a defensible RFP draft with scoring rubric and weighted criteria.
- [xRFI (drafter)](https://procureai.tech/.well-known/skills.json#xrfi-drafter): Draft a 20–40 question RFI for candidate suppliers — sits between xMatch and xRFP, qualifies suppliers before paying for a full RFP cycle.
- [xMatch (supplier shortlist)](https://procureai.tech/.well-known/skills.json#xmatch-supplier-shortlist): Ranked shortlist of 10–25 qualified suppliers from a category brief.
- [xScore (bid evaluation)](https://procureai.tech/.well-known/skills.json#xscore-bid-evaluation): Defensible weighted scoring matrix with sensitivity analysis for supplier RFP responses.

### Contracts & Compliance (4 skills)
- [xClause (risk-scoring)](https://procureai.tech/.well-known/skills.json#xclause-risk-scoring): Apply the buyer's playbook at clause level with risk scores and proposed redlines.
- [xContracts (summariser)](https://procureai.tech/.well-known/skills.json#xcontracts-summariser): Structured extraction from MSAs / SOWs / NDAs — obligations, SLAs, indemnity caps, exit clauses, renewal terms.
- [xNegotiate (coach)](https://procureai.tech/.well-known/skills.json#xnegotiate-coach): Rehearse a supplier negotiation against one persona at a time — CSM by default, switchable to CFO or lawyer. Returns BATNA / ZOPA estimates and three likely objections with counter-moves.
- [xRenew-radar](https://procureai.tech/.well-known/skills.json#xrenew-radar): 120-day renewal pipeline sorted by auto-renew exposure, price-escalator flags, and notice-period deadlines.

### Spend Analytics (4 skills)
- [xSpend-taxonomy](https://procureai.tech/.well-known/skills.json#xspend-taxonomy): Auto-classify a GL or AP export into a 4-level (L1–L4) spend taxonomy with three ranked consolidation plays. Feeds xMaverick / xSavings / xCash with the same row-level schema.
- [xMaverick (detector)](https://procureai.tech/.well-known/skills.json#xmaverick-detector): Surface off-contract, off-AVL, and split-PO patterns; ranks findings by exposure.
- [xSavings (narrative & CPO board QBR)](https://procureai.tech/.well-known/skills.json#xsavings-narrative): Three modes — per-deal savings narrative (FX-normalised, CFO-defensible), portfolio roll-up against plan, and a quarter-end CPO board QBR that stitches savings + supplier health (from xRisk-monitor digests) + renewals (from xRenew-radar) + market context (from xMarket-briefs) + 1–3 enforced asks for the board. The QBR is a provenance-tagged composition, not a regeneration.
- [xCash (working capital)](https://procureai.tech/.well-known/skills.json#xcash-working-capital): Model DPO uplift across the AP file — which vendors to standardise, working-capital release, per-vendor pushback risk. Dynamic-discount add-on opt-in via WACC / hurdle-rate input.

### Suppliers & Intel (4 skills)
- [xRisk-monitor](https://procureai.tech/.well-known/skills.json#xrisk-monitor): Supplier health across the lifecycle in three modes — continuous monitoring (news / filings / sanctions / ESG digest), rolling Health Card scorecard across 5 pillars (financial / operational / commercial / compliance / relationship) with composite grade and trend, and blameless incident postmortem after a real disruption (force majeure, single-source failure, cyber, sanctions) — `lessons` feed xClause playbook updates and `batna_implications` feed xNegotiate renewal prep.
- [xESG (scoring)](https://procureai.tech/.well-known/skills.json#xesg-scoring): Map vendor master to Scope-3 estimates, modern-slavery disclosures, and CSRD-aligned data.
- [xMarket-briefs](https://procureai.tech/.well-known/skills.json#xmarket-briefs): Weekly category briefs on commodities the buyer actually buys (steel, semis, freight, cloud, DRAM).
- [xVendorBench (shortlist benchmark)](https://procureai.tech/.well-known/skills.json#xvendorbench): Side-by-side benchmark on a 1–6 vendor shortlist. Two presets — `procurement-software` (S2P, contract AI, spend co-pilots) and `operational-supplier` (3PL, IT services, MRO, prof-services). Hands off cleanly from xMatch.

## Insights — field-tested writing for CPOs

Each piece is grounded in real engagements with numbers, trade-offs, and the next move. Free to read; no email gate on individual pieces.

- [The 12 procurement AI use-cases that pay back in one quarter](https://procureai.tech/insights/12-procurement-ai-use-cases.html): Three-tier ordering, the two patterns that didn't clear the bar, what to do Monday. 14-min playbook.
- [RFP automation honestly assessed](https://procureai.tech/insights/rfp-automation-honestly-assessed.html): Drafting half is mature (71% cycle-cut, real); scoring half isn't yet. Don't let a vendor bundle them. 11-min practitioner guide.
- [An AI measurement framework procurement can defend to the board](https://procureai.tech/insights/ai-measurement-framework-for-the-board.html): Four metrics with formulas, one-page audit-committee template. 9-min framework.
- [Pilot to production in 90 days](https://procureai.tech/insights/pilot-to-production-90-days.html): Week-by-week deployment plan with weekly-review template and five failure modes. 13-min playbook.
- [Opus 4.7 vs GPT-5.5 vs Gemini 3.1 Pro on procurement-real tasks](https://procureai.tech/insights/frontier-model-procurement-benchmark.html): Calibrated benchmark across RFP drafting, MSA extraction, Scope-3 inference, supplier risk synthesis. Plus the routing layer. 16-min benchmark.
- [Build vs buy vs fractional](https://procureai.tech/insights/build-vs-buy-vs-fractional.html): Three-question test, 8-row decision matrix, six vendor-pitch red flags. 8-min decision framework.

## About

- [Author — Martin Bacigal](https://www.linkedin.com/in/martinbacigal/): Founder of ProcureAI; Global Category Manager — IT at Nouryon; ex-Henkel Global Center of Excellence; $16M+ cumulative IT / SaaS / cybersecurity savings; built ProcurementGPT; led the global AI capability-building programme for 350+ procurement professionals.
- [About ProcureAI](https://procureai.tech/about.html): Team, the free Readiness Assessment, the ROI Calculator, the FAQ.
- [Contact](mailto:hello@procureai.tech): Direct line.

## Optional

- [Sitemap (XML)](https://procureai.tech/sitemap.xml): Full URL inventory.
- [Robots policy](https://procureai.tech/robots.txt): Explicit allowlist for GPTBot, ClaudeBot, anthropic-ai, Google-Extended, PerplexityBot, and the other major AI crawlers.
- [Extended LLM index](https://procureai.tech/llms-full.txt): Same as this file with skill descriptions, trigger phrases, and insight TL;DRs inlined for one-shot ingestion.
- [Plugin manifest (legacy)](https://procureai.tech/.well-known/ai-plugin.json): ai-plugin.json for agents that still parse it.
- [Agents manifest](https://procureai.tech/.well-known/agents.json): agents.json discovery file.