Skip to main content
Glama
Sign In

Procurement & Supplier Selection

by ai.saifs
OverviewDiscussionsAdmin

Choose & Justify a Supplier

procurement_supplier_selection
Read-only

Explained, audit-ready supplier selection and buying-quantity decisions in one call. Weights the user's criteria from pairwise importance answers and CHECKS THEM FOR CONTRADICTIONS (AHP consistency ratio; refuses and names the judgement to revisit), ranks suppliers by closeness to the ideal supplier respecting lower-is-better vs higher-is-better criteria (TOPSIS) and tests how robust the winner is to the weights, scores vendor reliability from delivery history (on-time and defect rates with 95% ranges, step-change and steady-drift trend checks, survival of problem-free delivery runs with the chance the next 5 deliveries are fine), works out whether quantity-discount price breaks are worth it (all-units discount EOQ, total annual cost), sorts spend into strategic / leverage / bottleneck / routine (Kraljic), and returns a decision record with an inputs hash and approver. Output starts with plain-language decision cards (CHOOSE, TOO CLOSE, FIX FIRST, STOP USING, ORDER, DO NOT BULK BUY, PROTECT, NEGOTIATE ...). Any subset of inputs works — e.g. only price_breaks for a bulk-discount question. Up to 1,000 delivery rows; larger files: procurement_get_engine.

Tool call template

This tool declares no required parameters. Add optional parameters from the input schema as needed.

{
  "id": 1,
  "jsonrpc": "2.0",
  "method": "tools/call",
  "params": {
    "arguments": {},
    "name": "procurement_supplier_selection"
  }
}

Input Schema

NameRequiredDescriptionDefault
as_ofNoDecision date YYYY-MM-DD (recorded; filters later deliveries).
criteriaNoCriterion names in order, e.g. [price, quality, delivery_days, service].
currencyNoISO code, e.g. INR. No FX conversion.USD
item_nameNoWhat is being bought, e.g. 'surgical gloves'.
suppliersNoOne object per shortlisted supplier: {name, <criterion>: number, ...}.
order_costNoCost of placing one order (admin, freight, inspection).
approved_byNoApprover name/role; leave empty until approved.
decision_idNoYour reference (file no., PO no.). Generated if absent.
spend_itemsNoOne row per item/category: {item, annual_spend, supply_risk (low/medium/high or 1-5)}.
cr_thresholdNoMax consistency ratio accepted.
holding_rateNoYearly cost of holding stock as a share of its value (0.15-0.30 typical).
price_breaksNoAll-units price slabs [{min_qty, unit_price}], or {supplier: [slabs]} to compare suppliers.
annual_demandNoUnits needed per year (for order sizing).
max_order_qtyNoOptional cap per order (storage, shelf life, budget).
on_time_floorNoOn-time target, e.g. 0.90.
tie_thresholdNoScore gap below which suppliers are tied.
currency_symbolNo
late_grace_daysNo
spend_thresholdNoOptional spend line between high and low (default: median).
criteria_weightsNoFallback when the user insists on fixed weights: {criterion: weight}. Not consistency-checked — prefer pairwise_comparisons.
delivery_historyNoOne row per delivery: supplier + (on_time yes/no OR promised_date + delivered_date); optional qty, rejected_qty, date. ERP headers accepted as-is.
criteria_directionsNocost = lower is better (price, lead time, defects); benefit = higher is better.
include_diagnosticsNo
procurement_contextNo'private' or 'public_tender' (government / PSU / GeM).private
pairwise_comparisonsNoImportance judgements on Saaty's 1-9 scale. Preferred form: a list of {a, b, value} meaning 'a is value times as important as b' (1 equal, 3 a bit more, 5 clearly more, 7 much more, 9 overwhelmingly more; fractions like 1/3 or "1/3" for the reverse). Every pair once. Also accepted: a square matrix (rows in `criteria` order), or a committee object {"member": [rows...], ...}.
min_deliveries_for_assessmentNo

Output Schema

NameRequiredDescriptionDefault
alertsNo
statusYes1 = ok/partial, 0 = nothing decidable
messageNo
metadataYes
decisionsYesPlain-language cards, priority order. Relay these first.
assumptionsNo
order_sizingNo
decision_recordNo
spend_portfolioNo
criteria_weightsNo
supplier_rankingNo
vendor_reliabilityNo
data_quality_reportNo

Schema Changelog

Changes observed during successful MCP inspections.

  1. First observed

TDQS

A4.5/5.0
Behavior5/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

Annotations already establish read-only, non-destructive, closed-world behavior, and the description adds substantial context beyond them: AHP consistency checking with refusal behavior, TOPSIS ranking, robustness testing, delivery-reliability scoring, EOQ/price-break analysis, Kraljic sorting, and decision-record output. These behavioral details materially help an agent understand what the tool will do and when it may refuse.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness3/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The purpose is front-loaded, but the description is a dense, long paragraph packed with acronyms, parentheticals, and output-detail lists. It earns most of its length for a 26-parameter tool, but it would be easier to scan with structure and less repetition of details already covered by the output schema.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness5/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

For a complex, zero-required-parameter tool with many optional inputs, the description covers the major analytical modules, the partial-input flexibility, row limits, sibling routing, and output form. With an output schema present, it need not fully document return values, and the description still gives enough for correct invocation.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

Schema description coverage is high at 85%, so the schema carries most parameter documentation. The description still adds useful global semantics: any subset of inputs works, price_breaks alone can answer a bulk-discount question, delivery history is capped at 1,000 rows, and larger files should use the sibling tool.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description states a specific verb and resource: explained, audit-ready supplier selection and buying-quantity decisions in one call. It clearly distinguishes the tool from the only sibling by directing larger files to procurement_get_engine, so an agent can identify what this tool does without opening the schema.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

It explicitly says any subset of inputs works and gives a concrete example of using only price_breaks for a bulk-discount question, which helps an agent understand partial-use modes. It also names procurement_get_engine as the alternative for larger files, though it does not state broader when-not-to-use conditions.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

All Procurement & Supplier Selection toolsInspect
Try in Browser
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

Glama MCP Gateway

Add one secure layer between your agents and this server.

Resources