Back to browse
GitHub Repository

Agent-native API for governed grant-proposal workflows: donor-aware drafting, preflight, human-in-the-loop review, grounding checks, and export-ready evidence packs across EU/FCDO/USAID/World Bank and 40+ donors. MCP + HTTP. Not a chatbot. Infrastructure.

6 starsPython

GrantFlow (FastAPI and LangGraph) for donor-aligned NGO proposal drafts

by vassilbek·Feb 24, 2026·1 point·2 comments

AI Analysis

●●SolidSolve My ProblemNiche Gem

Stateful workflow cuts grant drafting overhead, but narrowly addresses one domain's pain.

Strengths
  • Strategy Pattern isolation of donor-specific logic reduces prompt tuning surface area.
  • Human-in-the-loop checkpoints with explicit state transitions prevent runaway generation.
  • Exports to .docx/.xlsx and supports RAG-ready ChromaDB namespaces for donor compliance rules.
Weaknesses
  • No evidence of live user traction or comparative benchmarks against manual drafting time.
  • Donor strategy coverage listed but not validated — unclear if specialized prompts actually improve outcomes.
Category
Target Audience

NGO program officers, grant writers, implementing organizations (DAI, Chemonics, Tetra Tech)

Similar To

Proposal writing automation (Proposify, PandaDoc) · Grant search/tracking tools (Grants.gov, GrantStation)

Post Description

Hi HN,

I’ve been building GrantFlow, an open-source drafting workflow engine for institutional grant proposals.

The problem: many NGOs and implementing organizations spend a huge amount of time/money translating solid program ideas into donor-specific, reviewable proposal artifacts before they can even get meaningful feedback internally.

A lot of that work is not “thinking through the intervention” — it’s reshaping the same idea into structured outputs (ToC, LogFrame, MEL framing), aligning language to donor expectations, and managing review cycles.

GrantFlow is my attempt to reduce that overhead.

It takes structured project inputs and produces donor-aligned draft artifacts through a stateful workflow with review checkpoints (human-in-the-loop), instead of a single “generate everything” prompt.

What it does today (MVP): - Donor-aware drafting strategies (specialized + generic donor coverage) - Human-in-the-loop checkpoints (pause / approve / resume) - Exportable artifacts (.docx / .xlsx / ZIP) - RAG-ready donor knowledge namespaces (ChromaDB) - FastAPI API for integration into internal tools - Optional API key auth - Optional SQLite persistence for jobs + HITL checkpoints

Tech stack: - FastAPI - LangGraph - Pydantic - ChromaDB (with local/in-memory fallback) - Python 3.11+

Recent work I finished before posting: - hardened CI + shell checks - public API response redaction - typed response models for status endpoints - sqlite-backed job/HITL stores + WAL/busy_timeout - protected PDF ingest endpoint (`POST /ingest`) - readiness endpoint (`GET /ready`)

Why I built it this way: - proposal work is iterative and review-heavy - compliance/rules matter, so workflow/state matters - teams need checkpoints and auditability, not just raw text generation

Who I think this may be useful for: - implementing organizations (e.g. firms managing donor-funded programs) - NGOs and local partners - civic-tech / govtech teams building internal proposal tooling - consultants who standardize drafting workflows across donors

Happy to answer questions, especially around workflow design / HITL / donor strategy modeling.

Similar Projects

Nonprofit Results-Based Management logic model skill for OpenClaw

This skill automates the tedious parts of writing program logic models — it outputs a 5-level results chain, an if/then Theory of Change with assumptions, SMART indicators, SDG mapping and a monitoring plan. That feature set is exactly what M&E teams and grant writers want, but the public face is rough (ClawHub shows "Skill not found"), so the project needs clearer example outputs, ready-to-copy indicator templates, and better onboarding to move from useful hobby to everyday tool.

Niche GemSolve My Problem
vassilbek
104mo ago
AI/MLMid

We built AI to help bidding teams – mentioned by Kunal Bahl on ET Now

Focused on construction and engineering bids, ContraVault pulls RFPs into a workflow that auto-extracts requirements, highlights compliance risks, and generates draft proposal language plus go/no-go summaries. The vertical focus is smart — specialized models can beat generic RFP tools — but the site's bold '90% time savings' claim needs case studies or demo data to convince buyers in a crowded enterprise market.

Niche GemSolve My Problem
tanmayjuneja8
204mo ago
Social●●Solid

Donor reporting dashboard with one-click PPTX export (React/TS)

The one-click PowerPoint export is the real selling point — it packages KPIs, timelines and qualitative stories into slides using pptxgenjs, which actually solves the tedious manual rebuild nonprofits hate. The repo shows care (type-safe indicator calculations, Vite chunking, tests), but to go beyond an MVP it needs robust templating, spreadsheet/connectors and a reusable data model for different orgs and donor formats.

Niche GemSolve My ProblemSlick
vassilbek
104mo ago