Reciprocity & Single-ATO Overview — Executive Brief

One authorization, many agencies — with signed, self-verifying evidence per agency

Reciprocity & Single-ATO Overview — Executive Brief

The Single-ATO Reciprocity Model answers a specific federal problem: when one platform (for example, an OPM-operated HRIT service) is consumed by many agencies, do you re-authorize it N times, or once? The Federal Modernization Program’s canon ([the program decision]) is unambiguous: one System Security Plan, one controlling Authority to Operate, covering all consuming agencies, with per-agency differences expressed as configuration latitude rather than separate SSPs or POA&Ms.

This solution will have a single authoritative ATO covering all customers — agency differences are handled via configuration, not per-agency forks.

How reciprocity is recorded

Per FMP_140, each consuming agency receives a signed reciprocity record — not a new authorization, but a cryptographic acknowledgement that it operates under the controlling ATO. The Federal Modernization Program ships the full operational surface today:

Command What it does
fmp reciprocity onboard-agency Emits a signed (HMAC-SHA256) reciprocity record for a consuming agency, citing the controlling ATO and its legal basis
fmp reciprocity list-records Enumerates records and computes Active/Expired status against current time
fmp reciprocity verify Verifies a record’s HMAC signature
fmp conmon ato-cadence-check Enforces the SSP/reauthorization SLAs (draft ≤30d, final ≤45d, reauth ≥30d before expiry)

Each record names the controlling ATO, the consuming agency, the legal basis (e.g. FedRAMP Rev 5, OMB M-24-15, FISMA §3554), and the agency’s configuration elections — and is anchored as an ato-decision → reciprocity-record edge in the evidence graph (FMP_113 v1.2).

Self-verifying per-agency bundles

A consuming agency’s authorizing official should not need the Federal Modernization Program access — or even a network connection — to trust their evidence. The per-agency reciprocity bundle (FMP_140 §7) packages the reciprocity record, scoped component definition, assessment results, and a hashed manifest so the bundle self-verifies with nothing but the Python standard library. Tampering with any file breaks the manifest hash.

Configuration latitude is drift-governed

“Configuration-only differentiation” is not a loophole. Each agency’s elections are bounded by the SSP’s latitude table, and deviations surface as first-class drift findings: a configuration the SSP forbids is a DRIFT-AUTHZ P1; an unrecognized tenant principal is a DRIFT-IDENTITY P1. Eight KSIs (KSI-RECIP-001..008) continuously check the program’s health — legal-basis citation, controlling-ATO linkage, no silent expirations, SSP cadence (KSI-RECIP-005), the 30-day reauthorization window (KSI-RECIP-006), and bundle signature integrity.

The honest trade-off

Single-ATO reciprocity concentrates risk by design: a single authoritative ATO is the point of leverage and the single point of failure. If the controlling ATO lapses, every consuming agency is affected simultaneously. That is exactly why the cadence SLAs and the KSI-RECIP reauthorization indicators exist — the model trades distributed redundancy for centralized rigor, and the rigor has to be continuously proven.

Maturity snapshot

Capability Maturity
Reciprocity doctrine (the program decision) + model spec (FMP_140) SHIPPED (canon)
reciprocity CLI (onboard-agency / list-records / verify) SHIPPED
Signed record emitter + self-verifying bundle SHIPPED
conmon ato-cadence-check SLA validator SHIPPED
KSI-RECIP-001..008 indicators SHIPPED
Evidence graph ATO-decision / reciprocity-record nodes SHIPPED
End-to-end OSCAL SSP/POA&M reciprocity output profile TARGET (canon-defined; emission deferred per the program decision)

Where to go deeper

Leadership takeaway

The Federal Modernization Program turns reciprocity from a paperwork promise into a cryptographic, continuously monitored fact: one ATO, signed per-agency records that auditors can verify offline, configuration latitude bounded by drift detection, and eight standing indicators on program health. The remaining gap is the full OSCAL SSP/POA&M reciprocity output profile — defined in canon, deferred in emission — so treat the evidence and verification path as shipped and the generated authorization package as roadmap.

Back to top