| name | mcaf-testing |
| description | Add or update automated tests for a change using the repository’s verification rules in `AGENTS.md`. Use when implementing a feature, bugfix, refactor, or regression test; prefer stable integration/API/UI coverage and pull deeper test strategy from the bundled references. |
| compatibility | Requires the repository’s build and test tooling; uses commands from root or local `AGENTS.md`. |
MCAF: Testing
Trigger On
- implementing a feature or bugfix
- adding a regression test for a failure
- protecting a refactor with automated verification
Value
- produce a concrete project delta: code, docs, config, tests, CI, or review artifact
- reduce ambiguity through explicit planning, verification, and final validation skills
- leave reusable project context so future tasks are faster and safer
Do Not Use For
- repo-wide delivery policy with no test change
- documentation-only changes unless they alter executable verification
Inputs
- the nearest
AGENTS.md
- the changed behaviour and touched boundaries
- existing tests near the impacted code path
Quick Start
- Read the nearest
AGENTS.md and confirm scope and constraints.
- Run this skill's
Workflow through the Ralph Loop until outcomes are acceptable.
- Return the
Required Result Format with concrete artifacts and verification evidence.
Workflow
- Read the repo’s real verification commands from
AGENTS.md.
- Start with a failing test first when the change adds behaviour or fixes a bug.
- Start with the smallest meaningful test scope:
- new or changed tests
- related suite
- broader regressions
- When the stack is .NET, use the external
.NET skills from managedcode/dotnet-skills, use mcaf-dotnet as the orchestration skill when the task spans code, tests, and verification, and route framework mechanics through exactly one matching skill:
mcaf-dotnet-xunit
mcaf-dotnet-tunit
mcaf-dotnet-mstest
- Prefer integration, API, or UI coverage when behaviour crosses boundaries.
- Prove the user flow or caller-visible system flow, not just internal details.
- Add a regression test for every bug that can be captured reliably.
- If the stack is .NET and production code changed, do not stop at tests only. Finish with the repo-defined format and analyzer pass as well.
- Use deeper testing references only when the repo’s current strategy is unclear.
Deliver
- automated tests close to the changed behaviour
- verification results that match the repo’s real commands
Validate
- the new behaviour is covered at the right level
- the main user flow or caller-visible system flow is proven
- tests assert meaningful outcomes, not implementation trivia
- coverage expectations from
AGENTS.md are met, or the exception is documented
- the verification sequence matches
AGENTS.md
- for .NET changes, tests were not treated as a substitute for formatting or analyzer gates
- broader suites are run after there is something real to verify
Ralph Loop
Use the Ralph Loop for every task, including docs, architecture, testing, and tooling work.
- Brainstorm first (mandatory):
- analyze current state
- define the problem, target outcome, constraints, and risks
- generate options and think through trade-offs before committing
- capture the recommended direction and open questions
- Plan second (mandatory):
- write a detailed execution plan from the chosen direction
- list final validation skills to run at the end, with order and reason
- Execute one planned step and produce a concrete delta.
- Review the result and capture findings with actionable next fixes.
- Apply fixes in small batches and rerun the relevant checks or review steps.
- Update the plan after each iteration.
- Repeat until outcomes are acceptable or only explicit exceptions remain.
- If a dependency is missing, bootstrap it or return
status: not_applicable with explicit reason and fallback path.
Required Result Format
status: complete | clean | improved | configured | not_applicable | blocked
plan: concise plan and current iteration step
actions_taken: concrete changes made
validation_skills: final skills run, or skipped with reasons
verification: commands, checks, or review evidence summary
remaining: top unresolved items or none
For setup-only requests with no execution, return status: configured and exact next commands.
Load References
- read
references/test-planning.md first
- open
references/automated-testing.md for deeper strategy and trade-offs
- for broader .NET implementation flow, use the external
mcaf-dotnet skill from managedcode/dotnet-skills
- for .NET framework-specific mechanics, use exactly one external skill from managedcode/dotnet-skills:
mcaf-dotnet-xunit, mcaf-dotnet-tunit, or mcaf-dotnet-mstest
Example Requests
- "Add tests for this bugfix."
- "Protect this refactor with regression coverage."
- "Choose the right test level for this API change."