/rfp-responder

name: rfp-responder
description: "Use when an RFP, RFI, RFQ or security or vendor questionnaire arrives. Parses mandatory vs weighted requirements with rfp_parser.py, drafts proof-point answers with response_drafter.py and estimates win chance with winrate_predictor.py."
context: fork
license: MIT
metadata:
  version: 2.8.0
  author: claude-code-skills
  tags: [commercial, rfp, rfi, rfq, shipley, win-theme, proof-points, structured-response, bid-management]
  compatible_tools: [claude-code, codex-cli, cursor, antigravity, opencode, gemini-cli]

rfp-responder

Purpose

Help Bid Managers, Proposal Leads, and Directors of Sales answer five questions at the response-strategy moment:

1. **What is this RFP actually asking?** (parse sections, tag every requirement MANDATORY / WEIGHTED / NICE-TO-HAVE, extract scoring criteria, surface deadlines and format constraints) 2. **What is our true fit?** (proof-point matrix per requirement: STRONG / PARTIAL / GAP, each backed by a verifiable source — case study, certification, customer quote, technical attestation, benchmark) 3. **What is our win-theme strategy?** (Shipley method: 3-5 themes that ladder up across requirements, not generic value-prop bullets) 4. **What is our realistic winrate?** (Shipley-derived factor model: fit, incumbent, relationship strength, decision-criteria alignment, late-entry, competitor count, deal size — produces estimate + confidence band) 5. **Should we bid?** (deterministic verdict: BID / PARTNER-BID / NO-BID with named factors driving the call)

The skill surfaces GAPs explicitly. Leadership decides whether to close them, partner around them, or no-bid. **It never invents claims.**

When to use

• A 30+ page RFP / RFI / RFQ has landed with a 7-14 day response deadline
• A security questionnaire (SIG, CAIQ, custom-buyer) needs structured Q&A — not prose
• The team is preparing a bid / no-bid review and needs a defensible winrate estimate
• Sales Engineering has a proof-point library but no system to map proofs to requirements
• Leadership wants to see fit % (STRONG / PARTIAL / GAP) before committing pursuit budget
• A late-entry opportunity needs honest assessment of the relationship deficit

**Do not use for:**

• Free-form proposal narrative authoring → `business-growth/contract-and-proposal-writer`
• Contract redline AFTER award → `c-level-advisor/general-counsel-advisor`
• Marketing collateral / category content → `marketing-skill/*`
• Discount approval on the awarded deal → `commercial/deal-desk`
• Pricing-model design for a new product → `commercial/pricing-strategist`

Workflow

Step 1 — Parse the RFP

Drop the RFP markdown / text into `scripts/rfp_parser.py`. Output: structured JSON listing every requirement, tagged MANDATORY / WEIGHTED / NICE-TO-HAVE based on cue words (must / shall = MANDATORY; should / weighted scoring numbers = WEIGHTED; may / preferred / desired = NICE-TO-HAVE). Captures section structure, scoring criteria if disclosed, deadline, submission format constraints.

python scripts/rfp_parser.py --input rfp.md --output json > parsed.json

Step 2 — Score fit per requirement

Fill `assets/rfp_intake_template.md` with your proof-point library (each proof tagged with type + verifiable source + which requirement-tags it covers) and proposed win-themes. Feed parsed RFP + intake into `scripts/response_drafter.py`. Output: proof-point matrix per requirement with STRONG / PARTIAL / GAP, win-theme injection, GAP audit.

python scripts/response_drafter.py --input draft_input.json --output markdown > matrix.md

**Hard rule:** GAP requirements are surfaced, never invented around. Leadership reads the GAP audit and decides: close the gap, partner-bid, or no-bid.

Step 3 — Apply win-theme strategy

Shipley method: 3-5 themes that span requirements. Each theme answers "why us over the incumbent / competitor on the criteria the buyer named." `response_drafter.py` shows which themes thread through which requirements — a theme appearing in <2 requirements is decorative, not strategic, and gets flagged.

Step 4 — Estimate winrate

Feed deal context (fit %, incumbent strength, relationship, decision-criteria alignment, late-entry, competitor count, deal size vs. average) into `scripts/winrate_predictor.py`. Output: Shipley-derived estimate 0-100% + confidence band + factor breakdown + BID / PARTNER-BID / NO-BID verdict.

python scripts/winrate_predictor.py --input deal_context.json --profile enterprise-software --output markdown

**No-bid threshold:** estimate < 20% triggers automatic no-bid recommendation.

Step 5 — Decide

Take parsed RFP + proof-point matrix + GAP audit + winrate estimate into the go / no-go review. Skill does not commit pursuit budget — leadership does.

Scripts

• `scripts/rfp_parser.py` — section + requirement extractor (regex + cue-word heuristics, stdlib only)
• `scripts/response_drafter.py` — proof-point matrix + win-theme injection + GAP audit
• `scripts/winrate_predictor.py` — Shipley-derived factor model + bid/no-bid verdict, industry-profile-tuned

All scripts: stdlib only (argparse, json, sys, pathlib, re, collections, statistics). `--help` and `--sample` work on all three.

References

• `references/shipley_method_canon.md` — Shipley Proposal Guide v6, Shipley Capture Guide, APMP BoK, Tom Sant, Tom Searcy + Henry DeVries, Strategic Proposals research, Larry Newman
• `references/rfp_strategy_canon.md` — FAR, GSA, Forrester, Gartner, Bain, McKinsey, B2B International on RFP win-rates and buyer behavior
• `references/rfp_anti_patterns.md` — Shipley failure modes, APMP cases, Strategic Proposals research, federal loss reviews, MIT Sloan, Bain commercial-discipline, Gartner

Assumptions

• **The RFP is the ground truth.** If the buyer asked it, answer it — in the order they asked, in the format they specified. Re-organizing for narrative flow is for proposals, not RFPs.
• **Proof points must be verifiable.**