- name
SKILL.md
O arquivo da skill permanece no repositório do publicador.
Cole o comando copiado em um terminal. Ele instala este SKILL.md nesse agente.
npx skills add tt-a1i/archify --skill test-value -a cursor -yCole o comando copiado em um terminal. Ele instala este SKILL.md nesse agente.
npx skills add tt-a1i/archify --skill test-value -a claude-code -yCole o comando copiado em um terminal. Ele instala este SKILL.md nesse agente.
npx skills add tt-a1i/archify --skill test-value -a codex -yCole o comando copiado em um terminal. Ele instala este SKILL.md nesse agente.
npx skills add tt-a1i/archify --skill test-value -a grok -yEnvie a mensagem copiada em um chat do Grok Bot. O Bot busca este SKILL.md no GitHub e o adiciona às skills que todos os seus Bots compartilham.
Install this skill: https://github.com/tt-a1i/archify/tree/HEAD/.agents/skills/test-valueDo publicador
Protect outcomes worth preserving at a cost proportionate to their risk. Test counts, coverage percentages and green runs are evidence, not substitutes for product judgment. When an AI writes both implementation and tests, check whether they share the same mistaken assumption.
Preserve the user's existing authorization and operational boundaries.
Read decision-examples.md when dealing with automatic-layout snapshots, shared matrices, source assertions, asynchronous waits, test doubles, benchmarks or coverage gaps.
Explain what must remain true using user requirements, public contracts, compatibility commitments or a real failure. The current implementation and its own generated output cannot, by themselves, establish the correct expectation.
For tests being added or changed, answer three questions in the reasoning or a short change explanation:
Prefer observable output, state, data integrity and failure behavior. Private-function tests, snapshots and static checks can also protect real boundaries; their form alone does not determine their value.
Adding a test for every edit is not the objective. Pure copy changes, mechanical edits and changes already covered effectively may need no new test. When the user authorizes removing a product restriction, update the tests too; an old assertion does not veto an approved product change.
Preserve effective protection against data loss or corruption, permission and trust-boundary failures, promised compatibility, essential accessibility failures, resource leaks and broken recovery. Simplifying these checks still requires evidence of surviving protection. Bulk skips, retries, wider tolerances or refreshed snapshots must not conceal regressions. Changes to product commitments or mandatory release-platform gates must be within the user's authorized scope.
Use evidence proportionate to the risk instead of creating another test suite for every test:
Run destructive probes in temporary copies or controlled fixtures, preserving existing user changes. An unavailable environment is a validation gap; simulated evidence cannot stand in for a real browser, operating system or distribution package.
Read the project's actual test commands, CI selection rules and current changes before selecting the smallest set of checks that covers the risk. Affected behavior determines whether integration, browser, platform or complete verification is needed.
Confirm that the intended tests actually executed, including filters, collection and skips. A successful command with no relevant tests is not evidence that the behavior passed. Zero selection can be appropriate for changes with no affected tests, but establish that reason instead of inferring it from a green exit code.
Reuse results whose revision identity, inputs, environment and coverage remain applicable, and disclose what was reused. Expand or repeat verification when a new change, failure or unresolved risk invalidates evidence. Narrow failures for diagnosis instead of repeatedly launching the complete suite.
Claim speedups only from measurements. Distinguish case time, file time, a parallel pipeline's critical path and total compute. Attribute nested log timings correctly: a parent suite's duration is not its final child's duration. Control runtime, platform, inputs, cache state and competing work when comparing measurements.
When real use reveals a valuable new testing principle or a flaw in this guidance, you may refine this skill in place, within the active environment's permissions and the user's instructions. This is optional maintenance, not a required step in every task; keep it proportionate and preserve the original task's priority.
Before changing the skill, establish:
Prefer revising or consolidating an existing principle to appending another rule. Put cross-cutting principles in this entrypoint and conditional examples in the relevant reference. Keep additions in English, omit private incident details and transient measurements, and preserve authorization and protection boundaries. A lesson does not authorize broader repository changes, publication or synchronization to other installations.
Review the resulting guidance for contradictions and unnecessary obligations; check links and metadata if affected. Exercise a realistic decision scenario when the change alters behavior. If the skill is unavailable or read-only, propose the exact improvement and explain that it was not saved.
Whenever you update the skill, tell the user in the final response: link the changed file, summarize the principle added or revised, and explain the evidence or reason. Distinguish a saved change from a proposal. No update is needed when the existing guidance already explains the lesson.
Briefly report the protected behavior, test changes and rationale, executed or reused evidence, skips and intentional coverage tradeoffs. Report timing changes only with comparable measurements. Ordinary development does not need an additional governance report, scoring system or CI policy.
For audits, account for investigated candidates as changed, retained or awaiting evidence. A round can end when its high-value candidates are addressed and remaining candidates lack evidence or cost more than they save. The objective is neither endless deletion nor growing process. Keep focused, complete-suite, visual, platform and release acceptance claims distinct.
Improve test value and feedback cost when auditing slow, flaky, redundant or brittle tests; choosing tests during implementation; or reviewing test changes before a commit or PR. Keep ordinary impleme
---name: test-valuedescription: Improve test value and feedback cost when auditing slow, flaky, redundant or brittle tests; choosing tests…---# .agents/skills/test-value/SKILL.mdVisão geral
Improve test value and feedback cost when auditing slow, flaky, redundant or brittle tests; choosing tests during implementation; or reviewing test changes before a commit or PR. Keep ordinary implementation and pre-commit use scoped to the affected behavior, not a repository-wide audit.
Perguntas frequentes
Baixe .agents/skills/test-value/SKILL.md do publicador ou copie o caminho e abra o repositório. O PlugMCP lista a skill e não hospeda o arquivo.
O publicador o mantém em https://github.com/tt-a1i/archify.
Claude, Cursor, Codex e Grok Build conseguem ler um SKILL.md que você adiciona a partir do repositório do publicador. O Grok Bot o instala quando você envia em um chat o link da pasta da skill no GitHub.