IDE Software Prior Testing Summary

Status: Reviewer summary Owner: BionicLoop engineering Last updated: 2026-06-06 10:50 EDT

Purpose

Summarize the current body of prior software testing that supports the investigational baseline, without requiring reviewers to start from the full RTM or detailed STP/STR library.

Current Evidence Shape

The current software package already has broad engineering evidence across:

unit tests for core algorithm-host, runtime policy, alerting, persistence, and safety behavior
app-side unit tests for Home, meal, BG, CGM, pump, and alert workflows
targeted UI tests for critical operator flows
STP ownership documents for algorithm, automation, simulation, hardware, and alert drill lanes
controlled trace chain linking risk, requirements, design, verification rows, and review evidence

Latest Candidate-Baseline Regression Snapshot

The current candidate software baseline 963d717 includes a focused regression pass for masked fallback pump-delta recovery, same-pod continuity replay gating, no-replay continuation when pod continuity is not proven, and repeated no-active-pod alerting:

Evidence Area	Result
Full `BionicLoopCore` Swift package tests	Passed, `178` tests
Full app unit target `BionicLoopTests`	Passed
Focused fallback exposure / pump-delta coordinator suites	Passed
Focused masked fallback maintenance suite	Passed
Focused alert center and pump alert suites	Passed
Swift lint report mode	`0` serious findings; baseline warning-level length/complexity findings remain
Swift format report mode	`0/247` files require formatting
Xcode static analysis	Passed, exit code `0`
`git diff --check`	Clean

This regression snapshot supports the current candidate baseline but remains development/handoff evidence until promoted or rerun into the formal freeze evidence lane.

What This Means For IDE

For this packet, the useful claim is:

there is already a substantial prior engineering test body supporting the accepted local software baseline
the remaining work is not to invent new verification structure, but to trim and promote the right subset into the formal frozen IDE evidence set

Areas With Strong Existing Development Coverage

runtime wake-cause and degraded-mode behavior
masked fallback arm/refresh/reconnect recovery behavior, including same-pod pump-delta missed-step algorithm replay with unavailable CGM input and no-replay same-session continuation when pod continuity is not proven
meal announcement and manual BG workflow handling
alert normalization, acknowledgement, and notification behavior
no-active-pod alerting with repeat safety-critical notification attempts while unresolved
clinical settings gating and applied-configuration behavior
major Home/pump/CGM workflow rendering and state transitions

What Still Needs Freeze-Time Promotion Or Audit

RTM rows still marked In progress, Partial, or Pending
final audit of which rows are truly claimed in the IDE packet
formal promotion/rerun of specific high-risk or in-scope rows, including TV-SEC-001
final baseline SHA capture for the frozen evidence set

Reviewer Guidance

This packet should be read as:

evidence that the software package has meaningful prior engineering verification depth already
not yet a claim that every detailed RTM row is formally closed for the frozen IDE baseline

The freeze-time task is to carry forward only the evidence needed for the accepted packet scope.