Local-first MCP for building pitch decks and assessing startup funding readiness from on-device materials.
Copy the AI prompt to install this server into Claude Code, Cursor, or another agent β or use 1-click editor setup below.
This server is confirmed live β we successfully called its tools/list endpoint directly (see the verified badge above). We haven't yet sandbox-tested the stdio install command below specifically, which is a separate, ongoing check.
π‘ Paste the JSON block into your client's configuration file under mcpServers, then restart the application.
Inspect callable tools, capabilities, and parameters exposed to AI agents by Obra Cfo.
deck_blueprintReturn the framework for a fundable pitch deck at your stage: the slide structure with an evidence bar for each, the research to run first (as paste-in prompts for your Claude), and the artifacts to produce before the deck exists (business plan, pricing, go-to-market). Use this when you have no deck yet or a weak one. It reads your Obra CTO report if present. Then build the deck and ask the CFO to rank it.
prepare_assessmentRead the founder's funding materials (deck, plan, financials, one-pager) from a local path and return their contents with a rigorous assessment rubric. YOU, the host model, then assess them (separate facts from assumptions, extract the thesis, rate each dimension with evidence and confidence, find the weakest assumption, run the six-persona committee) and call score_funding_readiness with the structured result. Reads the Obra CTO report if present so technical claims rest on verified evidence. Materials stay local.
score_funding_readinessTurn your structured assessment (from prepare_assessment) into a stage-calibrated funding-readiness dashboard: investment score, confidence, risk register, the facts/assumptions/opinions ledger, your weakest assumption, what must be true, the unknowns, and the investment committee. Reads the Obra CTO report for technical risk.
The obra-cfo MCP server supports two related workflows: building a pitch and ranking an existing funding case. For an incomplete or weak deck, deck_blueprint returns a stage-specific slide structure, an evidence bar for each slide, research prompts, and a list of supporting artifacts to prepare, such as a business plan, pricing work, and go-to-market material.
For an existing fundraise, prepare_assessment reads local materials and supplies a rubric for the host model to apply. Supported inputs include a deck, business plan, financials, and one-pager. The model is expected to distinguish facts from assumptions, identify the central thesis, score dimensions with evidence and confidence, locate the weakest load-bearing assumption, and conduct a six-persona investment committee review.
score_funding_readiness formats that structured assessment as a stage-calibrated dashboard. The result can include an investment score, assessment confidence, risk categories, a facts/assumptions/opinions ledger, the weakest assumption, required conditions, unknowns, and committee perspectives.
The obra-cfo MCP server runs locally and makes no network calls. Materials remain on the machine where the server runs. The MCP tools return frameworks, source contents, and formatting logic; the host model performs the substantive assessment using the supplied rubric and then sends its structured result to the scoring tool.
The server can also consume the Obra CTO report from .obra/cto-report.json when that file is present. This report supplies technical evidence for the funding assessment, including technical-risk context. If the report is absent, technical claims are treated as assertions rather than verified evidence, and the workflow recommends running Obra CTO first.
A typical sequence is to request a deck blueprint if the pitch is not ready, prepare an assessment after the materials exist, and then submit the host model's structured assessment for scoring.
The README provides a Claude Desktop or Claude Code MCP configuration using the npm package:
After configuration, prompts can request a seed-stage pitch deck blueprint or ask the model to assess materials in a local directory. The project can also be run from source by cloning the repository, installing dependencies, building it, and configuring Node to run dist/index.js from an absolute path.
deck_blueprint: Returns a fundable-deck framework, evidence requirements, research prompts, and pre-deck artifacts for the founder's stage.prepare_assessment: Reads specified local funding materials and returns their contents with an assessment rubric.score_funding_readiness: Produces a structured, stage-calibrated funding-readiness dashboard from the host model's assessment..obra/cto-report.json when available and uses it as technical evidence.The obra-cfo MCP server is not a deck generator. Its blueprint provides structure and prompts, while the founder and host model create and evaluate the actual materials. It also does not guarantee funding or replace an investor's diligence process.
The assessment depends on the host model correctly applying the rubric and returning the expected structured result. The server itself does not upload materials or make external network requests. The free preview does not include the deeper playbook, expanded committee, or remediation templates described in the README's paid-community context.
Always refer to the official documentation for the most accurate and up-to-date information.
Factual signals from GitHub, npm, and our automated checks β not a rating.
No reviews yet β be the first to share how this listing worked for you.
Showcase your server listing on GitHub or your project documentation. Embed this dynamic SVG badge to highlight official listing status and live engagement.
[](https://allmcps.com/mcp/obra-cfo)<a href="https://allmcps.com/mcp/obra-cfo"><img src="https://allmcps.com/api/badge/obra-cfo?style=directory" alt="Obra Cfo on AllMCPs" /></a>