Simulink Explain Missing Coverage
Use this skill when the user asks to summarize, or asks why Simulink Coverage...
Install
npx skills add matlab/simulink-agentic-toolkit --skill simulink-explain-missing-coverage -g -a claude-code -y- Repository
- matlab/simulink-agentic-toolkit1.3k
- Category
- Design
- Updated
- 6d ago
- License
- See repository
- Context cost
- Mediumabout 2,981 tokens
- Runs scripts
- No, instructions only
What it does
Use this skill when the user asks to summarize, or asks why Simulink Coverage objectives are missing, unsatisfied, or uncovered on a coverage result they already have, why specific decision, condition, MCDC, relational-boundary, or saturation/overflow outcomes weren't exercised, whether uncovered logic is dead code, or how to close the gap (a test, a coverage filter, or a design change). It reads an existing cvdata, .cvt file, sim-with-coverage output, or Simulink Test result; it never runs cvsim to collect coverage.
Explaining Missing Simulink Coverage
Explain why coverage objectives are missing on a coverage result the user already
has, and what to do about each one, across all objective types (decision,
condition, MCDC, relational boundary, saturation/overflow, lookup-table execution).
Works from an existing cvdata / cvdatagroup, a .cvt file, a
Simulink.SimulationOutput with coverage on, or a Simulink Test result set.
It keeps two questions separate:
- Is this outcome reachable?, for decision/condition/MCDC objectives, answered by formal proof (SLDV dead-logic detection) joined to the coverage result deterministically. A provably-dead outcome is a fact.
- If reachable, why wasn't it tested, and if dead, was that intended?, answered by the agent, reasoning over an input trace (Model Slicer) plus model structure. This is judgment, always presented as a recommendation.
Requires: MATLAB R2023a+, Simulink Coverage, Simulink Design Verifier (dead-logic detection), Simulink Check / Model Slicer (input tracing). The skill never collects coverage, it reads a result that already exists. When SLDV or the Model Slicer is unavailable, the affected step degrades honestly (see Guardrails), never fabricating evidence.
When to Use
- User asks why an objective, of any metric, or a specific true/false outcome is missing, unsatisfied, or uncovered.
- User asks whether uncovered logic is dead code (unreachable) or just untested.
- User asks how to close a coverage gap on a result they already collected, what test to add, what to filter, or what to change.
- User needs audit-ready triage: provably-dead vs. still-to-test, each traced to its inputs.
- User points at an existing coverage artifact and asks to explain it.
