Testing¶

pi-vault-mind is tested at three levels: automated unit tests (Node test runner), manual end-to-end verification (a human runs through the watcher pipeline), and agent / HITL test plans (an LLM runs the test).

Documents in this section¶

Automated tests

Test plans for the three personas (Agent, Human, Human-in-the-Loop), the regression checklist, and the smoke-test one-liner.
Manual E2E

The full manual end-to-end verification: file change → watcher detection → subagent dispatch → results in LanceDB + Vault. 4 test scenarios.

Quick smoke test¶

The fastest way to verify a fresh install is healthy:

pi -e npm:pi-vault-mind
/wiki init
/wiki validate

If this sequence completes without errors, the extension is healthy.

Running the automated tests¶

pnpm test              # runs node --test dist/test/index.test.js

Tests use a mock ExtensionAPI and verify:

Tool/command/event registration counts
Config scaffolding correctness
Collection CRUD (append, query, dedup)
Export formats (JSON, CSV, Markdown)
Injector activation on prompt matching
Stats accuracy

Each test runs in a temp directory to avoid side effects.

Regression checklist¶

Run before each release:

Check	How
Build	`pnpm build` → zero errors
Tests	`pnpm test` → all pass
Version	`cat package.json \\| grep version` matches planned release
Changelog	`CHANGELOG.md` has entry for this version
Skills manifest	`skills/vault-mind/SKILL.md` has valid frontmatter
No stale refs	No `/qmd-init`, `/qmd-validate`, or `/wiki-*` in docs