Stop Paying 8% Commissions. Win More Grants Predictably.
Traditional freelance commission writing forfeits $25,000–$50,000 per award and violates federal Uniform Guidance (2 CFR 200.430). Ironclad Grants replaces predatory contingency fees with an end-to-end AI Grant Operating System: ProPublica 990-PF hydration, zero-hallucination pgvector RAG drafting on 24GB+ Billama GPUs, 3-persona Red Team peer review, and full post-award SF-425 compliance.
Funder & 990-PF Match
Grants.gov sync, ProPublica 990-PF curves & 5-axis SVG Fit Radar (A+ to F)
HNSW Knowledge Vault
Private 501(c)(3), 990 & biosketch indexing in Supabase RLS + SeaweedFS
AI Copilot & Red Team
8-section drafting on 24GB+ Billama GPUs + 3-persona 1–9 peer review
Pre-Flight & ZIP Binder
50-char filename sanitizer, 1" margins, SVG Logic Models & XML cover
Post-Award 2 CFR 200 OS
OMB SF-425 compiler, 200.414 MTDC solver, 200.430 Effort & SHA-256 log
Inside the Ironclad Grant Operating System
Explore every architectural pillar below or jump directly into the live module on portal.ironcladgrants.com.
Funder Intelligence & 990-PF Engine
Grants.gov, Candid & ProPublica 990-PF Auto-Hydration
Stop guessing ask amounts. Ironclad synchronizes federal opportunities (Grants.gov, DOE, USDA, NIH, NSF, EPA) with private foundation tax filings to compute mathematical fit before you write a single word.
Enter any Foundation EIN to pull 3-year Part XV grant schedules, asset totals, and median award sizes automatically.
Scores Mission Alignment, Legal Eligibility, Funding Size Fit, Cost-Share Feasibility, and Competitiveness with letter grades (A+ to F).
Computes the 25th–75th percentile giving curve and recommends the optimal first-time vs. repeat grantee request amount.
Verify Your Mission Savings & 2 CFR 200 Math
Test our statutory engines right here, then launch the full workspace on portal.ironcladgrants.com to generate your SF-424A budget and narrative justification.
Forfeited from project capital. Impermissible under 2 CFR 200.430 & AFP Code of Ethics.
Includes 45 pooled AI drafts/month, full HNSW RAG Knowledge Base & Red Team review.
100% retained inside your organization with zero federal audit clawback liability.
Why Funder Databases & Generic Chatbots Fail Federal Audits
Legacy databases abandon you once you find an RFP, generic AI tools invent disqualifying statistics, and commission writers violate federal law.
| Capability / Dimension | Contingency Writers (5%–10% Fee) | Legacy Databases (Candid / FDO) | Generic AI (ChatGPT / Claude) | Ironclad Grants 🛡️ (Grant OS) |
|---|---|---|---|---|
| Primary Operating Scope | Manual freelance writing for 5%–10% of award | Funder search & 990 lookup only (Candid / FDO) | Ungrounded chat prompt generation (ChatGPT / Claude) | End-to-End Grant OS: Research, RAG Drafting, Red Team, Pre-Flight & Post-Award |
| 2 CFR 200.430 & AFP Ethical Compliance | Impermissible contingent fee — high Single Audit clawback risk | Permissible subscription (no writing or compliance engine) | Violates NIH NOT-OD-23-147 / NSF PAPPG without certified disclosure | 100% permissible SaaS operational cost (2 CFR 200.454) + AI Disclosure Generator |
| Cost on a $500,000 Grant Award | $25,000 – $50,000 forfeited from mission capital | $2,000 – $5,000/yr (search database only) | $240/yr + 60 hrs manual formatting & high disqualification risk | $2,844/yr (3 Growth Seats) — Retain +$37,156+ for your mission |
| Anti-Hallucination Grounding | Manual copy-paste from old Word files | N/A | None — invents citations, statistics, and institutional history | Private pgvector HNSW RAG with inline source-level verification alerts |
| Federal Financial & Post-Award OS | Writer disappears immediately after submission | None | Cannot compute MTDC exclusions or SF-425 Box 10 math | Built-in SF-424A, 2 CFR 200.414 MTDC solver, 200.430 Effort PAR & SF-425 compiler |
| Consortium & Subrecipient Collaboration | Chaotic email attachments & version collisions | Basic shared saved searches | Single-user chat threads | Isolated Subrecipient Guest Portal, SAM.gov UEI check, RBAC & SHA-256 audit log |
Tailored Impact Across Your Organization
Freelance consultants demand 5%–10% contingency fees ($25k–$50k on a $500k award) that violate 2 CFR 200.430 and trigger Single Audit clawbacks.
Flat, predictable per-seat subscription ($79/seat/mo). 100% permissible under 2 CFR 200.454 with a tamper-evident SHA-256 audit trail.
Open Standards, SDKs & Activity Walkthroughs
Everything you need to configure multi-tenant workspaces, validate YAML proposal schemas, or deploy the Kubernetes Helm chart.
Platform & SDK Documentation
HNSW RAG architecture, 2 CFR 200 math specs, @tekromancy/grant_utils API reference, and live YAML Frontmatter Playground.
Explore DocsStep-by-Step User Guides
Actionable walkthroughs for 990-PF hydration, AI Copilot drafting, Red Team reviews, SF-425 reporting, and CLI workflows.
Read WalkthroughsInstant SeaweedFS digital vault delivery and hardware tokens for SAM.gov / Login.gov.
Latest Platform Release Notes
Continuous delivery to portal.ironcladgrants.com
Introduces unified Organization Sign In / Sign Up authentication workflows, completes the migration of the web portal to Next.js 14 App Router file-based routing, upgrades pgvector to HNSW indexing, and ships automated Grafana observability ConfigMaps.
Integrates @billama/sdk to route grant narrative drafting through 24GB+ decentralized Billama GPU nodes with token metering, wires FreeIPA LDAP and SeaweedFS object storage, and launches the /shop Grant Compliance Store.
Delivers the 6-workstream Phase 2 Expansion: Public Grant Readiness Diagnostic Scorecard, Multi-Modal Logic Model & WBS SVG Generator, Official OMB SF-424/SF-424A Form Engine, 2 CFR 200 Subpart E Fiscal Audit Engine, and Real-Time PQL CRM Webhooks.