Skip to content

readiness.md

Latest commit

 

History

History
109 lines (85 loc) · 4.49 KB

readiness.md

File metadata and controls

109 lines (85 loc) · 4.49 KB
Raw

Operational Readiness

Spanda's Operational Readiness Engine answers one question:

Can this robot safely perform this mission right now?

It composes existing platform gates — hardware verification, capability verification, health checks, connectivity validation, safety rules, and mission requirements — into a single weighted readiness score and go/no-go report.

Quick start

spanda readiness examples/showcase/readiness/rover.sd
spanda readiness examples/showcase/readiness/rover.sd --target RoverV1
spanda readiness examples/showcase/readiness/rover.sd --runtime
spanda readiness examples/showcase/readiness/rover.sd --runtime --inject-health-faults
spanda readiness examples/showcase/readiness/rover.sd --json
spanda readiness examples/showcase/readiness/rover.sd --agent-json
spanda readiness examples/showcase/policy/warehouse.sd --policy WarehousePolicy
spanda readiness examples/showcase/secure_boot/rover.sd --agent Rover@Jetson
spanda check examples/showcase/readiness/rover.sd --readiness-json --json
spanda demo readiness

Runtime and agent readiness

  • --runtime — evaluate against live hardware monitor signals (health checks use runtime fault/event state).
  • --inject-health-faults — simulate degraded sensors for what-if analysis (pairs with --runtime).
  • GET /v1/readiness on deploy and fleet agents; query runtime=true and inject_health_faults=true for on-device evaluation.
  • POST /v1/program on deploy and fleet agents to upload .sd source for live readiness.
  • Remote CLI: spanda deploy agent readiness <Robot@Hardware> and spanda fleet agent readiness <RobotName>.
  • --agent-json on spanda readiness — same JSON envelope as agent GET /v1/readiness (for scripts, CI, and local parity checks).
  • spanda check --readiness-json — merges operational readiness diagnostics with recovery-policy diagnostics (recovery:policy, recovery:approval, recovery:fleet) and continuity-policy diagnostics (continuity:policy, continuity:fleet, continuity:approval, continuity:handoff). See verification-diagnostics.md.
  • Web playground — switch to Operations view in packages/web for local scoring or agent fetch.

Example output:

Mission Ready: YES
Score: 92/100

Issues:
* LTE signal weak
* Camera calibration due in 5 days
* Battery below recommended threshold

Readiness factors

Factor Source
Hardware spanda verify / hardware profiles
Capabilities Capability registry + minimum hardware
Health health_check declarations
Connectivity Connectivity policy + hardware
Safety Minimum capabilities + kill switches
Battery Mission duration vs battery budget
Storage / Compute Resource budgets (when declared)
Packages / Providers Traceability matrix
Mission Requirements mission { requires capabilities [...] }
Operational policy spanda readiness --policy <name> — weighted OperationalPolicy factor and policy violations as readiness issues

Operational policy integration

When --policy <name> is set, readiness evaluation merges verify-time operational policy rules into the report:

  • Adds an OperationalPolicy factor to the weighted score (weight 8)
  • Maps policy violations to readiness issues (High for errors, Medium for warnings)
  • Sets mission_ready = false when the policy fails

Use the same flag with deployment gates: spanda deploy gate <file.sd> --operational-policy <name>.

See policy-engine.md.

Types

  • ReadinessStatus — Ready, Degraded, NotReady, Unknown
  • ReadinessReport — full evaluation with score and issues
  • ReadinessScore — weighted total and per-factor breakdown
  • ReadinessIssue — severity, factor, message, suggested action
  • ReadinessPolicy — minimum score threshold and factor weights

Related commands

Command Purpose
spanda readiness <file.sd> Unified readiness evaluation
spanda verify mission <file.sd> Mission achievability check
spanda fleet readiness <file.sd> Fleet-level readiness
spanda twin readiness <file.sd> Physical vs digital twin drift
spanda audit <file.sd> Autonomous safety auditor

Crate

Rust API: spanda-readiness (evaluate_readiness, ReadinessReport, …)

See also: Mission Verification, Fleet Readiness, Safety Reporting.