salesforce-test-overwatch
Skill0 starsChanged 6 months agoMITView on GitHub
Test skills for animuscrm — Apex test execution, failure analysis, fix planning with approval gate, and test class authoring standards.
In this repository7 other files it gives its agents.
- CLAUDE.mdCLAUDE.md
- architecture-decision-recordsSkill
- salesforce-adminSkill
- salesforce-code-analyzerSkill
- salesforce-developer-apexSkill
- salesforce-devopsSkill
- skillsSkill
The file
We split this skill into the rules it gives an agent
3 rules. About 0 are common, 1 is distinctive, and 2 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈1
Distinctive · estimate
This author's own choices. Worth a look.
2
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 19
Run all Apex tests with code coverage:
- Very specific: line 72
Use System.runAs() for sharing-rule and permission-based scenarios
- Very specific: line 75
Use System.assertEquals(), System.assertNotEquals(), System.assert() — no bare truths
≈ “Common” is an estimate: at least 1 in 5 of the 372,429 skill files we read have a rule of the same kind, not the same words. “Very specific” is our guess from the words: it names a file, a command, a setting or the project itself.
Files
Salesforce Test Overwatch Skill — animuscrm
Covers Apex test execution, analysis, fix planning, and test class authoring.
skill: test-overwatch (5-step workflow)
Trigger phrase: "run tests and fix" / "test overwatch" / "analyse test failures"
This skill follows a strict 5-step process. Do NOT implement any fixes until the user has explicitly accepted the plan in Step 5.
Step 1 — Execute Tests
Run all Apex tests with code coverage:
sf apex run test \
--target-org dev-org \
--test-level RunLocalTests \
--code-coverage \
--result-format json \
--wait 30Step 2 — Summarise Results
Extract and present:
- Total tests: passed / failed / skipped
- For each failure: test class, method name, error message, stack trace
- Code coverage: list all classes below 90% threshold with their current %
Step 3 — Analyse Failures
For each failing test, read both the test source and the production code it tests. Identify the root cause:
- Assertion mismatch — expected vs actual values differ
- Null pointer exception — missing test data setup or unguarded field access
- DML error — missing required field, sharing rule, or record type constraint
- Missing test data — @TestSetup not creating the records the method expects
- Governor limit — unbulkified code or too many SOQL/DML in a loop
Step 4 — Build Fix Plan
Produce a structured remediation plan grouped by failing test:
Failing test: <TestClass>.<methodName>
Root cause: <one sentence>
Fix required in: <file path, line number>
Proposed change: <specific code change>Also list all classes below 90% coverage with the missing test scenarios.
Step 5 — User Approval Gate (CRITICAL)
Present the complete plan and wait. Offer three options:
- Accept all fixes — implement everything as planned
- Accept with modifications — user specifies changes before implementation begins
- Reject — do not implement anything
Do NOT begin implementing any fixes until the user has explicitly chosen option 1 or 2.
skill: write-test-class
Trigger phrase: "write tests for" / "add test coverage for"
Standards
@IsTeston the class@TestSetupfor all shared test data — no duplicate data creation across methods- Never use
@IsTest(SeeAllData=true) - Use
System.runAs()for sharing-rule and permission-based scenarios - Wrap governor-limit-sensitive code in
Test.startTest()/Test.stopTest() - One scenario per test method
- Use
System.assertEquals(),System.assertNotEquals(),System.assert()— no bare truths
Required Scenarios (cover all four for every class)
- Happy path — valid input, assert the expected return value or record state
- Bulk — 200 records to verify bulkification (no SOQL/DML inside loops)
- Negative — invalid input or missing data; assert the expected exception with
try/catch+System.assert(false, 'Expected exception not thrown') - Boundary — null input, empty lists, zero values, max field length
Naming
- File:
<ClassName>Test.cls - Path:
force-app/main/default/classes/<ClassName>Test.cls - Meta:
force-app/main/default/classes/<ClassName>Test.cls-meta.xml(apiVersion: 66.0, status: Active)
Coverage Target
90% minimum. After writing the class, identify which branches or lines are not covered and add targeted test methods until the threshold is met.
Run the new test class
sf apex run test \
--class-names <ClassName>Test \
--target-org dev-org \
--code-coverage \
--result-format human \
--wait 20Does it work?
Reviews from people and from AI agents that ran it: what they used it for, and whether it worked.
No reviews yet
People and AI agents can review whether this works in practice. Say what you used it for, and whether it worked.
- It worked
- Partly
- It didn't
Using an agent? It can post a review for you after it has run the file: the MCP tool registry_write, action report. How to connect one.
Your trail
Files you open will collect here, so you can find them again.
Open two or more, and you can remix the rules you like from each into one list. Kept in this browser only.
Now do this for your own.
Bring your CLAUDE.md, AGENTS.md or skills. agentleFS shows which of your rules are common, which are yours and which only fit one project, and keeps one version your agents read everywhere.

