Trust Center

Capital Command runs on a live, multi-tenant Supabase database with real user authentication and Row Level Security enforced at the database level — a department's data is only ever visible to signed-in members of that department. This is checked by an automated suite of more than 900 database security tests, run before releases, that test the database directly rather than just the application. Departments can sign up on their own, upload documents to private storage, and receive in-app notifications. Capital Command is an early-stage product; see “What's not built yet” below.

Departments own the organizational data, documents, and materials they submit. Wood Capital LLC owns Capital Command's software, code, architecture, workflows, and brand, subject to applicable third-party and open-source licenses. Generated outputs (reports, funding strategies, board materials) are intended for broad use by the department that generated them.

Capital Command uses AI in these features: the AI Fire Chief Assistant; grant application narrative drafting, guided application interviews and narrative review; annual report and board meeting drafting; SOP/SOG, apparatus purchasing and recruitment drafting tools; the knowledge assistant; and the Mutual Aid Agreement Assistant. Each uses only information the requesting user is authorized to see — the department's own information, or the text of a mutual aid agreement the department is party to, including documents it uploads — is processed by Anthropic as a data processor, and is never used to train AI models. AI results stay private to the requesting department, and AI never signs, approves or sends an agreement. AI output is labeled AI-suggested and unverified until a person reviews it, and AI never invents a department fact or a regulatory requirement as verified. Funding matching, eligibility checks, financial calculations, readiness checks and apparatus scoring are deterministic, rule-based computations against recorded facts — never AI-generated.

Capital Command runs a scheduled, daily ingestion pipeline against public funding sources — federal programs through Grants.gov's public API, fire-service grant listings, and curated state programs. Discovered opportunities are stored with full source attribution and start out marked “unverified,” never presented as confirmed until a human reviews them. This product uses the Grants.gov API but is not endorsed or certified by the U.S. Department of Health and Human Services. Every funding program record carries a verification status and, where applicable, a source URL and last-checked date. Coverage does not yet include every federal, state, local, foundation or corporate program — broader coverage is ongoing work.

Capital Command is a research and organizational tool. It never guarantees eligibility for a funding program, an award outcome, or an award amount — every eligibility check states its basis and flags what's unknown rather than assuming the best case. Capital Command does not provide legal, financial, tax, or grant-writing advice; a department should independently confirm anything important with the actual funding source and its own advisors before relying on it for a real decision or submission.

Being direct about gaps matters more than sounding finished: email notifications for new funding matches and deadlines are not turned on yet (notifications currently appear in the app only); data export and deletion are handled on request rather than self-service; and eligibility matching against automatically-discovered opportunities is still coarse — it avoids guessing at a rejection when information is missing, so many discovered opportunities show as a potential match pending closer review, not a confirmed strong fit.

Draft Terms of Service and Privacy Policy are available for review. Both are marked as drafts pending attorney review and are not yet in effect.

Security or data questions about a pilot engagement should go through your Capital Command point of contact directly.