RFQ Generator v1.1 — Cowork Plugin

The Alternative to Tradogram. Turn a 6-hour RFQ-to-PO workflow into 45 minutes included with your subscription in licensing.

Installation

1. Download the pf-rfq-generator.plugin file
2. Open Claude Desktop and navigate to Settings > Plugins
3. Click Install Plugin and select the downloaded .plugin file
4. The plugin will be installed and available immediately

Note: All data stays local on your machine. No external API calls or cloud storage required.

Why This Exists

Procurement teams at SMBs spend ~$2,340-$4,500 per year on Tradogram to generate RFQs, collect supplier quotes, compare bids, and create purchase orders. This workflow is 80% document generation—exactly what Claude does best. RFQ Generator eliminates the Tradogram subscription with an AI-powered Cowork plugin that handles the full pipeline.

What you pay: Included with your subscription What Tradogram costs: $2,340-$4,500/yr (Pro to Premium) Savings: $2,340-$4,500 annually per SMB

Quick Start (5 Steps)

1. Setup — Run /rfq:rfq-setup once per project folder. Provide your company name and address.
2. Intake — Run /rfq:rfq-intake. Describe your line items conversationally or upload a CSV. The plugin extracts and validates specs automatically.
3. Generate RFQ — Run /rfq:rfq-generate. Get a professional DOCX RFQ ready to send to suppliers.
4. Parse Quotes — Suppliers respond. Upload their CSV/XLSX files and run /rfq:rfq-parse-quotes. The plugin normalizes all quote formats intelligently.
5. Compare & Approve — Run /rfq:rfq-compare to see weighted scoring (price, lead time, quality). Run /rfq:rfq-approve to select the winning supplier. Run /rfq:rfq-po to generate the final PO.

Total time: 45 minutes instead of 6 hours.

Commands

Command	What It Does	When to Run
/rfq:rfq-pipeline	PRIMARY ENTRY — runs all 6 stages end-to-end	First time, or standard workflow
/rfq:rfq-setup	Initialize workspace, venv, templates, config	Once per project folder
/rfq:rfq-intake	Collect RFQ details and line items	Start a new RFQ
/rfq:rfq-generate	Create professional DOCX RFQ	After intake
/rfq:rfq-parse-quotes	Import and normalize supplier quotes	When quotes arrive
/rfq:rfq-compare	Score suppliers, recommend winner	After parsing quotes
/rfq:rfq-approve	Record explicit supplier approval	Before PO generation
/rfq:rfq-po	Generate final Purchase Order DOCX	Last step
/rfq:rfq-status	Check pipeline progress	Anytime

How It Works

The plugin orchestrates a 6-stage procurement workflow:

1. Setup — Create project folders (.rfq/, inbound/, outbound/, processing/, archive/), set up Python environment, validate templates
2. Intake — Collect RFQ specs conversationally or from CSV; AI extracts and validates
3. Generate — Load template, substitute company/RFQ data, populate line-item table, save DOCX
4. Parse Quotes — Auto-detect encoding/delimiter, map non-standard supplier columns, flag anomalies
5. Compare — Weighted scoring (price 50%, lead_time 30%, quality 20%), AI-powered anomaly detection, recommendation
6. Approve — Explicit decision gate; record supplier selection with timestamp
7. PO — Cross-check specs vs. quote, generate final Purchase Order DOCX

All intermediate results are checkpointed in .rfq/project.json, allowing resume from any stage.

AI-Powered Features

Smart Spec Extraction (rfq-intake) Describe what you need in plain language. The plugin structures it into a professional RFQ format, asking clarifying questions for vague specs. No TBD placeholders allowed.

Intelligent Quote Scoring (rfq-compare) Beyond raw price, the plugin identifies bid anomalies (suspiciously low prices may indicate quality risk), flags missing specs, and highlights favorable terms you might overlook. Clear recommendation with reasoning.

Context-Aware PO Generation (rfq-po) Cross-references your original RFQ specs against the winning supplier's quote. Flags discrepancies (spec changes, qty mismatches, delivery term shifts) so nothing is missed between quote acceptance and formal order.

Automatic Quote Parsing (rfq-parse-quotes) Suppliers send quotes in different formats with different column names and encodings. The plugin auto-detects encoding, delimiters, and maps non-standard headers to a normalized schema — no manual copy-pasting required.

Centralized Procurement Knowledge Base (rfq-kb) References authoritative standards including UCC Article 2 (§2-205 firm offers, §2-207 battle of the forms, §2-314/§2-315 implied warranties), FAR Parts 13/15 (evaluation factors, price analysis thresholds), and ICC Incoterms® 2020 (all 11 rules with risk/cost transfer points).

Procurement Validation Checklists (rfq-kb) Five built-in checklists validate RFQ completeness per UCC §2-205, supplier quote completeness, quote comparison fairness per FAR §15.304, PO legal sufficiency per UCC §2-207, and procurement approval authority thresholds.

Incoterms-Aware Quote Comparison (rfq-parse-quotes, rfq-compare) Detects when suppliers quote different Incoterms (e.g., EXW vs DDP) and flags that unit prices are not directly comparable without freight/duty normalization per ICC Incoterms® 2020.

Early Payment Discount Analysis (rfq-parse-quotes, rfq-compare) Identifies and calculates the annualized return on early payment discounts (e.g., 2/10 Net 30 = 37.24% annualized) and surfaces them as favorable factors during comparison.

UCC-Compliant PO Generation (rfq-po) Generated POs include Terms Precedence clause per UCC §2-207 battle of the forms, warranty references per UCC §2-314/§2-315, inspection/acceptance period per UCC §2-606, and legal disclaimer.

FAR-Aligned Evaluation Methodology (rfq-compare, rfq-generate) Evaluation criteria are disclosed in the RFQ per FAR §15.304, scoring documents strengths/weaknesses/risks per FAR §15.305, and price analysis flags unrealistically low offers per FAR §15.404.

Approval Authority Awareness (rfq-approve) Flags when order values exceed typical delegation thresholds (e.g., >$10K may need procurement manager, >$50K may need director/VP) per rfq-kb procurement approval authority matrix.

What It's an Alternative To

Vendor	Annual Cost	pf-rfq-generator
Tradogram Pro	$195/mo ($2,340/yr)	Included
Tradogram Premium	$375/mo ($4,500/yr)	Included
Procurify	$5,000-$15,000/yr	Included (but no ERP sync)
Precoro	$6,000-$12,000/yr	Included (but no QuickBooks sync)
Team Procure	$3,000/yr	Included

RFQ Generator may significantly reduce annual costs for typical SMBs.

Key Differentiators

• Included with your subscription — No subscription required; runs in Cowork
• Data stays local — Supplier pricing and terms never touch a vendor's cloud
• Customizable templates — Edit the RFQ and PO DOCX templates in your Cowork project for any industry
• AI-powered analysis — Price anomalies, missing specs, and risk flags are identified automatically
• No feature bloat — Focused on the 5-step RFQ-to-PO workflow SMBs actually use; designed to displace ~68% of Tradogram's core value

Feature Comparison

Feature	pf-rfq-generator	Tradogram	Procurify
RFQ Document Generation	✅ Full	✅ Full	✅ Full
Quote Comparison Matrix	✅ Full (XLSX + DOCX)	✅ Full	✅ Full
Purchase Order Generation	✅ Full	✅ Full	✅ Full
CSV/XLSX Quote Parsing	✅ Full	⚠️ Partial (manual)	✅ Full
AI Anomaly Detection	✅ Full (bid risk flags)	❌ None	❌ None
Approval Gate	✅ Full (rfq-approve)	⚠️ Partial	✅ Full
Persistent Supplier Database	❌ None	✅ Full	✅ Full
Direct Email to Suppliers	❌ None	✅ Full	✅ Full
ERP/QB Integration	❌ None	⚠️ Partial	✅ Full
Multi-User Approvals	❌ None	⚠️ Partial	✅ Full
Spend Analytics	❌ None	✅ Full	✅ Full
Data Privacy (Local)	✅ Full	❌ Cloud	❌ Cloud
Procurement Knowledge Base	✅ Full (UCC, FAR, Incoterms, Kraljic, TCO)	❌ None	❌ None
Incoterms-Aware Comparison	✅ Full (flags non-comparable quotes)	❌ None	❌ None
Early Payment Discount Analysis	✅ Full (annualized return calc)	❌ None	❌ None
UCC-Compliant PO Terms	✅ Full (§2-207 Terms Precedence)	⚠️ Basic	⚠️ Basic
Procurement Validation Checklists	✅ Full (5 checklists)	❌ None	❌ None
Approval Authority Thresholds	✅ Full (value-based flagging)	⚠️ Partial	✅ Full
Cost (Annual)	included with your subscription	$2,340-$4,500	$5,000-$15,000

Estimated Cost per Use

Disclaimer: Token estimates are approximate and based on typical usage patterns measured from skill prompt sizes. Actual costs vary with input data size, conversation length, and complexity. Estimates use Claude Sonnet 4.6 pricing ($3/1M input, $15/1M output). Cowork and Claude Desktop subscription users (Pro/Max/Team) are not charged per-token — these estimates apply only to direct Anthropic API usage. Running stages individually in fresh sessions uses fewer input tokens than running the full pipeline sequentially, because pipeline mode accumulates conversation history across stages.

Per skill (run individually in a fresh session):

Stage	Skill Prompt	User Input	Total Input	Output	Est. Cost
rfq-po	~4.2K	~800	~8.2K	~4.2K	~$0.09
rfq-compare	~4.0K	~800	~7.9K	~4.0K	~$0.08
rfq-approve	~2.5K	~800	~6.3K	~3.0K	~$0.06
rfq-parse-quotes	~3.3K	~2.0K	~8.2K	~2.0K	~$0.05
rfq-kb	~5.9K	~800	~9.5K	~5.9K	~$0.12
rfq-generate	~3.0K	~800	~6.8K	~6.0K	~$0.11
rfq-intake	~3.5K	~800	~7.3K	~3.5K	~$0.08
Standalone total			~54.1K	~28.6K	~$0.59

Full pipeline (all stages in one session — context accumulates):

Stage	Base Input	+ History	Total Input	Output	Est. Cost
rfq-po	~9.0K	0	~9.0K	~4.2K	~$0.09
rfq-compare	~8.8K	~5.0K	~13.8K	~4.0K	~$0.10
rfq-approve	~7.3K	~9.8K	~17.1K	~3.0K	~$0.10
rfq-parse-quotes	~9.3K	~13.6K	~22.9K	~2.0K	~$0.10
rfq-kb	~10.8K	~17.6K	~28.3K	~5.9K	~$0.17
rfq-generate	~7.9K	~24.3K	~32.2K	~6.0K	~$0.19
rfq-intake	~8.4K	~31.1K	~39.5K	~3.5K	~$0.17
Pipeline total			~162.8K	~28.6K	~$0.92

Running the full pipeline once typically costs $0.64–$1.19 in API tokens (Claude Sonnet 4.6).

Known Limitations & Workarounds

Limitation	Impact	Workaround
No persistent supplier database	Supplier contact, certifications, history not tracked session-to-session	Setup creates supplier-list.csv template; user populates once and imports for each RFQ cycle
No live email integration	Cannot send RFQs directly to supplier inboxes	Generate supplier email template + contact export CSV; user sends via Gmail/Outlook
No ERP/accounting sync	POs cannot be auto-pushed to QuickBooks/NetSuite	Export PO as PDF/CSV; user manually imports in 5 minutes (standard SMB workflow)
No multi-user approval workflow	Single-user tool; cannot route POs to managers	Share DOCX/PDF via Google Drive for async team review before approval
No spend analytics dashboard	Cannot track trends month-over-month in live dashboard	Accumulate per-cycle XLSX reports; build custom pivot tables in Excel
Session-based only	No background processing between sessions	Checkpoint file (.rfq/project.json) tracks progress; resume from any stage

Context & Performance Guide

The plugin uses a Python virtual environment (.rfq/venv/) with three dependencies:

• python-docx — DOCX generation and template substitution
• openpyxl — XLSX comparison matrix generation
• chardet — Automatic CSV/XLSX encoding detection

Typical RFQ workflows (10-20 items, 3-5 suppliers) use ~40-50% of Claude's context window. No chunking required.

For large RFQs (50+ items, 10+ suppliers), the plugin guides batching: process suppliers in groups of 5, generate separate comparison matrices, then manually merge results.

Requirements

• Cowork — Cowork integration installed in your browser
• Python 3.8+ — For the .rfq/venv/ virtual environment (auto-created by setup)
• Folder access — A writable folder where RFQ project files will be stored
• Column-mapped CSV/XLSX — Supplier quotes should have recognizable headers (Supplier, Unit Price, Lead Time, etc.); flexible mapping handles variations

Important Disclaimers

• AI-Generated Content: This plugin uses AI (LLM) technology which can produce inaccurate or incomplete outputs. All content should be treated as a starting point and reviewed for accuracy before use.
• Not Professional Advice: Outputs do not constitute legal, financial, tax, medical, or other professional advice. Consult qualified professionals before making decisions based on generated content.
• No Compliance Guarantee: References to industry standards, regulations, or guidelines are for informational purposes only. This plugin does not guarantee compliance with any law or regulation. Users are responsible for verifying all outputs meet their specific regulatory requirements.
• No Endorsement or Affiliation: Mention of third-party products, standards, or organizations does not imply endorsement, partnership, or certification by those entities.