rfp-response-builder
Skill1 starsChanged 25 days agoApache-2.0View on GitHub
Draft answers to RFPs, RFIs and security questionnaires (SIG, CAIQ, custom xlsx or docx) and prepare proposals, from the user's own answer library and documents. Every answer cites a source and gets a green, amber or red confidence. Unsourced answers go to an SME, never guessed. Includes a consistency check across answers and a bid or no-bid view. Use when the user shares a questionnaire or RFP, or asks for a proposal, or when presales-ai routes here.
1 other project uses a version of this.
Adopted by
Most recently taken up by
- 00PrabalK00/claude-skillstheir own version, first seen Mar 16, 2026
Other versions1 project uses a version of this.
Word for word · 0
None found.
Edited · 0
None found.
Wrote their own version · 1
- 00PrabalK00/claude-skills★ 1same name
How matches are found. Word for word: the exact same text. Edited: shares passages with this file (“99% same” means 99% of its five-word phrases also appear here; 10% or more counts). Their own version: similar meaning, or, weakest, the same name. We can’t tell which project had it first. Projects are counted among the 200 most-starred candidates, and mentions are links to the repository we found, up to 50 per site, so both counts are a floor.
In this repository13 other files it gives its agents.
- architecture-decision-recordSkill
- competitive-battle-cardsSkill
- creative-animated-svgSkill
- deal-qualification-deskSkill
- drawio-creative-diagramsSkill
- excalidraw-creative-diagramsSkill
- find-skillsSkill
- llm-finopsSkill
- markdown-craftSkill
- presales-aiSkill
- presales-meeting-prepSkill
- solution-design-explorerSkill
- tdr-authorSkill
The file
We split this skill into the rules it gives an agent
7 rules. About 0 are common, 2 are distinctive, and 5 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
5
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Very specific: line 8
Read deal-context.json first if it exists (see presales-ai/references/deal-context.md).
- Very specific: line 19
If there is none, build it from past responses and docs the user points to.
- Very specific: line 29
red: no match. Draft only from a document you can cite.
- Very specific: line 43
Write the answers into the buyer's file and format (xlsx skill for spreadsheets, docx skill for Word).
- Very specific: line 49
Use references/proposal-outline.md.
- Distinctive: line 55
Every answer is green with a source, amber with a source and a review flag, or red with an owner.
- Distinctive: line 57
No em-dashes or semicolons in the prose you wrote.
≈ “Common” is an estimate: at least 1 in 5 of the 350,495 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
RFP, RFI and security questionnaire responses
Read deal-context.json first if it exists (see presales-ai/references/deal-context.md).
Step 1: Split the questionnaire
python3 scripts/split_questionnaire.py <file.xlsx|.csv|.md|.txt> -o out/questions.jsonlIt finds the question column, keeps the section and row id, and skips headers and blanks. For docx or PDF, extract the text first (the docx or pdf skill) and pass the .txt.
Step 2: Load or build the answer library
answers.jsonl in the working folder, one entry per line. Schema: references/answer-library.md.
If there is none, build it from past responses and docs the user points to. Each entry needs
source and last_verified. Entries older than 12 months are stale.
Step 3: Match and draft
python3 scripts/match_answers.py out/questions.jsonl answers.jsonl -o out/drafts.jsonlIt scores each question against the library (token overlap) and sets confidence:
- green: a strong match with a fresh source. Reuse it, adapt the wording.
- amber: a partial match, or a stale source. Draft it and flag it for review.
- red: no match. Draft only from a document you can cite. If none, write
SME needed: <owner area>and leave the answer empty.
Then read each amber and red row and improve the draft from the user's documents. Cite the file and section. Never invent a control, certification, number or customer.
Step 4: Consistency check
python3 scripts/consistency_check.py out/drafts.jsonlFlags the same topic answered with different numbers or yes and no (for example uptime 99.9 vs 99.95, or "SSO: yes" vs "SSO: roadmap"). Fix every conflict before you continue.
Step 5: Write back and report
- Write the answers into the buyer's file and format (xlsx skill for spreadsheets, docx skill
for Word). Keep their columns. Add
ConfidenceandSourcecolumns only in an internal copy. - Report counts by colour, the list of SME questions grouped by owner, and conflicts fixed.
- Offer to add approved new answers back to
answers.jsonl.
Proposals
Use references/proposal-outline.md. Map each section to the buyer's evaluation criteria and
win themes, pull persona and competitor from deal-context.json, and run the same
cite-or-flag rule. Add a bid or no-bid box at the top: coverage % (green plus amber over
total), red count, deadline risk and fit score if qualification ran.
Done means
- Every answer is green with a source, amber with a source and a review flag, or red with an owner.
consistency_check.pyreports 0 conflicts.- No em-dashes or semicolons in the prose you wrote.
Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.