# Testing tool recommendation

Target platform, installed runner/version, system and date: {{target}}
Concrete testing gap and current evidence: {{gap}}

| Candidate | Capability | Official source/date/version | Local research path | Cost/prerequisites | Qualification |
| --- | --- | --- | --- | --- | --- |
| {{tool}} | {{gap_addressed}} | {{source}} | {{local_source_or_none}} | {{runtime_and_dependencies}} | {{researched_fixture_tested_target_verified}} |

Preferred candidate and alternatives: {{decision_with_rationale}}
Proposed installation/version pin: {{command_not_executed_by_research}}
Runner/CI/configuration integration location: {{integration}}
Expected improvement and measurable acceptance: {{benefit}}
Limitations and transfer assumptions: {{limits}}
Platform template and evidence adapter changes: {{template_adapter}}
Validation command and deliberate failing control: {{qualification}}
Authorization/status of installation: {{status}}

Do not transfer a published score or percentage to this target without evidence. Mutation, property, browser, contract,
lint and flake tooling each address different boundaries; no single tool certifies all tests. Research output proposes
tools rather than silently installing them.
