docmole / tests
CLAUDE.md18 starsChanged 10 months agoMITView on GitHub
Rule: One test file per source module. Don't create embedded.test.ts if tests belong in config.test.ts.
- Reads credentials
In this repositoryOne other file it gives its agents.
- CLAUDE.mdCLAUDE.md
The file
We split this CLAUDE.md into the rules it gives an agent
2 rules. About 0 are common, 2 are distinctive, and 0 are very specific to this project.
≈0
Common · estimate
Many files like this have a rule of this kind.
≈2
Distinctive · estimate
This author's own choices. Worth a look.
0
Very specific
Tied to their project's names, files or tools. Probably skip.
- Common
- Distinctive
- Very specific
- Distinctive: line 64
What bug would this catch?: If none, skip it.
- Distinctive: line 65
Is this testing MY code or a library?: Only test your code.
≈ “Common” is an estimate: at least 1 in 5 of the 26,980 CLAUDE.md 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
Test Guidelines
File Organization
tests/
├── config.test.ts # src/config/* (project config, loader)
├── mintlify-api.test.ts # src/backends/mintlify.ts
├── embedded.test.ts # src/backends/embedded/*
└── <module>.test.ts # One file per major moduleRule: One test file per source module. Don't create embedded.test.ts if tests belong in config.test.ts.
What Makes a Test Useful
Good Tests (Keep)
// 1. Integration with real systems (LanceDB, APIs)
test("persists data across instances", async () => {
const kb1 = await createKnowledge(path, embedder);
await kb1.addDocument({ name: "doc", content: "test" });
await kb1.close();
const kb2 = await createKnowledge(path, embedder);
expect(await kb2.countDocuments()).toBe(1); // Real persistence!
});
// 2. Business logic that can fail
test("deduplicates by source URL (max 2 chunks)", async () => {
// Tests actual deduplication logic, not just "it runs"
});
// 3. Error handling
test("handles errors gracefully", async () => {
const mock = { retrieve: async () => { throw new Error("fail"); } };
const result = await tool.execute({ query: "test" });
expect(result.success).toBe(false); // Doesn't crash!
});Bad Tests (Remove)
// 1. Testing hardcoded values
test("default model is gpt-4o-mini", () => {
expect(config.model).toBe("gpt-4o-mini"); // So what? If it changes, is that a bug?
});
// 2. String matching on generated text
test("instructions contain 'agent'", () => {
expect(instructions).toContain("agent"); // Superficial
});
// 3. Testing library behavior
test("LanceDB returns array", () => {
expect(Array.isArray(results)).toBe(true); // Test your code, not theirs
});Before Writing a Test, Ask
- What bug would this catch? If none, skip it.
- Is this testing MY code or a library? Only test your code.
- If this fails, is it actually broken? Changing defaults isn't a bug.
Integration Tests
// Skip if dependencies unavailable
const hasOpenAIKey = !!process.env.OPENAI_API_KEY;
describe.skipIf(!hasOpenAIKey)("OpenAI Integration", () => {
setDefaultTimeout(60_000); // API calls are slow
// ...
});Mocking
// Type-safe mocks
type MockKnowledge = Pick<EmbeddedKnowledge, "search">;
const mock: MockKnowledge = {
search: async () => [{ name: "doc", content: "test", metadata: {} }],
};
// Cast safely
const retriever = new Retriever(mock as unknown as EmbeddedKnowledge);Naming
// Describe WHAT it does, not HOW
test("deduplicates by source URL (max 2 chunks per URL)") // Good
test("filters results array with map and reduce") // Bad
// Test files match source
// src/backends/embedded/* → tests/embedded.test.ts
// src/config/* → tests/config.test.tsRunning Tests
bun test # All tests
bun test tests/embedded.test.ts # Single file
bun test --watch # Watch modeDoes 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.

