helpmetest-discover
Who you are: If
.helpmetest/SOUL.mdexists in this project, read it before starting — it defines your character and shapes how you work.
No MCP? The CLI has full feature parity — use
helpmetest <command>instead of MCP tools. See the CLI reference.
QA Roamer
Explores a website to understand what it does and who uses it.
Prerequisites
Before creating artifacts, load the authentication guidance and tag schema. These define how to set up test users and how to tag scenarios consistently.
Call these first:
how_to({ type: "authentication_state_management" })
how_to({ type: "tag_schema" })
Workflow
More from help-me-test/skills
helpmetest
Single entry point for all HelpMeTest QA work. Dispatches to a mode based on the first argument: agent (Tasks-artifact harness, base discipline), tdd (write/fix tests — default for code-work tasks), discover (map site into Features), fix-tests (repair failing), coverage (gap analysis), regression (change-targeted run), validate (test quality review), report (read-only project health diagnosis), proxy (tunnel localhost), api-testing (API-level RF tests), ui-review (visual walkthrough), onboard (new project bootstrap). Usage: /helpmetest [mode] [task...]. Bare /helpmetest runs full QA (discover + tdd).
40tdd
Everything to do with tests on HelpMeTest. Use when: writing tests for a new feature, generating tests for an existing feature, fixing a broken test, debugging a failing test, tests broke after a UI change, tests are out of date after a refactor. Triggers on: 'write tests', 'generate tests', 'test is failing', 'fix tests', 'tests broke', 'implement X', 'add feature', 'fix bug', 'why does this test fail', 'tests are out of date'. If it involves HelpMeTest tests in any way, this is the skill.
36helpmetest-self-heal
Autonomous test maintenance agent. Monitors test failures and fixes them automatically. Always use this when tests start failing after a UI or code change — it's far more systematic than trying to fix tests manually one by one. Use when user mentions 'fix failing tests', 'heal tests', 'auto-fix', 'monitor test health', 'tests broke after deploy', or test suite has multiple failures needing systematic repair. Distinguishes fixable test issues (selector changes, timing) from real application bugs.
30helpmetest-debugger
When a test is broken and you need to know why, use this skill. It handles: a named test failing with an error message (element not found, selector not found, timeout), determining whether the failure is a bug in the app vs. a broken test, a test that passes locally but fails on CI, and multiple tests regressing after a deploy or PR merge. The distinguishing signal: the user has a failing test and wants root cause, not just a re-run. Do not use for: writing new tests, running test suites, exploring features, or reviewing test logic in the abstract.
26helpmetest-validator
Invoke this skill when a user shares test code and questions whether it actually works as intended — not to run or fix the test, but to evaluate whether the test has real value. Triggers on: \"is this test any good?\", \"would this catch a real bug?\", \"this test always passes — is that normal?\", \"review these tests before I commit\", or \"does this test verify anything meaningful?\". Also triggers when someone suspects a test is useless, wants a pre-commit quality gate, or is unsure if an auto-generated test is worth keeping. The core question this skill answers: \"Would this test fail if the feature broke?\" If not, the test gets rejected. Do NOT use for generating new tests, fixing failing tests, or exploring application features.
26helpmetest-test-generator
Use this skill when the user wants tests written for a specific feature or flow. Triggers on: \"write tests for X\", \"generate tests for checkout\", \"create tests for login\", \"add tests for registration\", \"we have scenarios — now write the tests\", or any request to produce automated test coverage for a known feature. Also triggers when discovery is done and the user is ready to move from documenting scenarios to actually testing them. Not for: exploring a site to discover what to test, judging whether an existing test is good, or debugging a failing test.
26