Spanda

Operational Governance Framework

Spanda’s Operational Governance Framework extends the platform with abstractions for how autonomous systems are designed, validated, deployed, operated, audited, and maintained — without becoming a generic GRC product or a repository of regulatory text.

Disclaimer: Spanda provides governance abstractions and validation mechanisms. It does not provide legal or regulatory advice.

Platform service

The spanda-governance crate (crates/spanda-governance/) implements:

Responsibility	Integration
Standards awareness	`StandardsProfileRef` via packages/plugins
Compliance validation	Composes with `spanda-compliance`, `spanda-readiness`, `spanda-trust`
Deployment governance	`DeploymentProfile` + entity metadata
Operational policies	`GovernancePolicyRef` (versioned, signed, auditable)
Certification tracking	Independent lifecycle from health
Risk assessment	`OperationalRisk` influencing decision/recovery
Audit support	`spanda-audit` + evidence export
Human accountability	`HumanAccountability` on missions/deployments

Architecture

spanda.governance.toml  →  ResolvedSystemConfig  →  EntityRegistry
                                                        ↓
                                              EntityGovernanceMeta
                                                        ↓
                    spanda-governance::evaluate_entity_governance
                                                        ↓
              CLI / REST / SDK / Control Center governance views

Governance does not duplicate readiness, assurance, trust, or health engines. It composes them through the unified entity model and validation orchestration.

Entity extensions

Every EntityRecord may expose optional governance via governance: Option<EntityGovernanceMeta>:

Autonomy level
Deployment profile
Operational maturity
Certification status
Risk level
Governance policies
Responsible owner
Standards profiles
Operational constraints

Configure system-wide defaults in spanda.governance.toml (auto-loaded when present, or via [config] governance = "spanda.governance.toml"). Override per-robot in device tree metadata.

CLI

spanda compliance check [--strict] [--entity <id>] [--json]
spanda governance validate [--strict] [--json]
spanda governance report [--json]
spanda certification list|inspect|report [entity-id] [--json]
spanda deployment profile [name] [--json]
spanda deployment verify [--strict] [--json]
spanda risk report [--json]

REST API

Method	Path	Description
GET	`/v1/governance`	Framework summary
GET	`/v1/compliance`	Compliance posture summary
POST	`/v1/compliance/check`	Run compliance check
POST	`/v1/governance/validate`	Validate governance configuration
GET	`/v1/certifications`	List entity certification records
GET	`/v1/deployment-profiles`	List or fetch deployment profile
GET	`/v1/risk`	Operational risk rollup
GET	`/v1/certifications/report`	Certification report
POST	`/v1/deployment/verify`	Deployment verification
GET	`/v1/governance/policies`	Policy assignments + audit
POST	`/v1/governance/policies/assign`	Assign/sign policy to entity
GET	`/v1/governance/audit`	Governance audit history
GET	`/v1/governance/accountability`	Responsible owners
PUT/POST	`/v1/governance/accountability`	Update owners / contacts / approval chain (entity overlay)
POST	`/v1/governance/policies/detach`	Detach a policy assignment

Live decision authorization (authorize_action / distributed decision engine) loads entity governance from the project config and entity overlay. High/life/mission-critical risk and manual/assisted autonomy require human approval for non-reflex actions; emergency stop and other safety actions remain permitted.

gRPC (proto 1.0.12)

RPC	REST parity
`GetGovernance`	`GET /v1/governance`
`GetCompliance`	`GET /v1/compliance`
`CheckCompliance`	`POST /v1/compliance/check`
`ValidateGovernance`	`POST /v1/governance/validate`
`ListCertifications`	`GET /v1/certifications`
`ListDeploymentProfiles`	`GET /v1/deployment-profiles`
`GetDeploymentProfile`	`GET /v1/deployment-profiles?name=`
`GetOperationalRisk`	`GET /v1/risk`

SDK

Rust, Python, and TypeScript SDKs expose dedicated clients:

GovernanceClient — framework summary, validate
ComplianceClient — summary, check
CertificationClient — list certifications
DeploymentProfileClient — list/get profiles
RiskClient — risk summary

Control Center

The Governance tab (Governance & config group) shows deployment profiles, compliance validation, certification status, operational risk, and accountability fields.

Smoke and Stable promotion

./scripts/operational_governance_smoke.sh
SPANDA_GOVERNANCE_SKIP_SOAK=1 ./scripts/operational_governance_stable_promotion_gate.sh

See stable-hardening-operational-governance.md.

This site is open source. Improve this page.