test-suite

Estrelas0

Forks0

Atualizado7 de junho de 2026 às 22:39

Delegate an ad-hoc test run to Codex and report results in chat (no artifact) — the cheap inner-loop step; /code-review is the checkpoint. Use on /test-suite or 'run the tests'.

Instalação

Instalar com Codex ou Claude Copie este prompt, cole no Codex, Claude ou outro assistente e deixe que ele revise a página da skill e instale para você.

Fonte

trevormil

trevormil/offline-first-sportsbook

Abrir repositório GitHub Ver repositórios do creator

Download

Ocupações relacionadasSOC

Baseado na classificação ocupacional SOC

Analistas de garantia de qualidade de software e testadoresInformática e Matemática·SOC 15-1253

SKILL.md

readonly

name	test-suite
description	Delegate an ad-hoc test run to Codex and report results in chat (no artifact) — the cheap inner-loop step; /code-review is the checkpoint. Use on /test-suite or 'run the tests'.

/test-suite — Run tests, report in chat

Ad-hoc test runs. Detects the runner per .agents/testing.md, runs the suite once, and reports results in chat. Writes no artifact.

This is the cheap inner-loop step: run it between commits while iterating to confirm tests stay green, then run /code-review at a checkpoint for the full six-axis review (which runs tests itself as its gate and embeds the result in the .reviews/ artifact). Use /test-suite when you want to know whether tests pass without paying for a full review.

Delegate to Codex

codex exec -C "$PWD" "/test-suite"

Run directly in Claude only if codex is unavailable or the user asks.

Process

Detect the runner per .agents/testing.md (package.json scripts.test → bun test/vitest/jest; pyproject → pytest; Cargo.toml → cargo test; go.mod → go test ./...; etc.). If multiple, prefer the one CI runs.
Install deps if needed (only if unresolved — no node_modules/.venv). Prefer bun install per global §5.
Run once — capture stdout+stderr, exit code, wall time. No retries.
Report in chat:

**Runner:** <bun test / pytest / ...>
**Command:** <exact command>
**Status:** pass | fail | partial | error
**Counts:** N passed, N failed, N skipped (total N)
**Duration:** X.Xs

<one-line summary; on failure, list test name + key error excerpt>

What NOT to do

Don't write artifact files. That's /code-review's job (the combined .reviews/ artifact).
Don't edit test files to make them pass. Failing tests become findings.
Don't mark pass when tests were skipped due to install failures — that's error.
Don't re-run until green. One run per invocation; flaky tests are findings worth a /ticket.

Activity

After the run, emit a feed event with the result:

# pass:
.claude/bin/activity tests-pass "Tests pass · <N> passed" "<runner> in <time>"
# fail:
.claude/bin/activity tests-fail "Tests fail · <F> failed / <N>" "<runner>"

Mais deste repositório

mesmo repositório

check

trevormil/offline-first-sportsbook

Run a scheduled cadence agent from an .agents/<kind>.md spec — isolated worktree, SHA-gated no-op on re-runs, then reports findings or proposes a ticket+MR. Use on /check <kind>, a scheduled-agent run, or wiring a launchd schedule.

2026-06-070

code-review

trevormil/offline-first-sportsbook

Delegate a GitHub/GitLab PR review to Codex: a deterministic preflight does recon, then Codex scores six axes into one artifact at .TerMinal/reviews/<pr>/<sha>.md (v2) or .reviews/<pr>/<sha>.md (legacy v1). Use on /code-review or 'review this PR'.

2026-06-070

emergency-fix

trevormil/offline-first-sportsbook

FORCE-MODE hotfix: smallest patch → commit → push DIRECTLY to main → follow-up ticket. Needs TERMINAL_FORCE_MAIN=1 (hook-enforced). Use on /emergency-fix or 'this is breaking prod' / 'hotfix' / 'we need to revert now'.

2026-06-070

factory

trevormil/offline-first-sportsbook

Continuous autonomous orchestrator — the no-handoff loop around /stacked-mr: reconcile (/merge-sync), build+batch-review a stack to the bar, optionally refill via discovery, repeat. Parks HITL, never merges to main. Use on /factory or 'work the backlog autonomously / continuously'.

2026-06-070

new-agent

trevormil/offline-first-sportsbook

Create or update a TerMinal agent from a natural-language request — writes a script-first .agents/<id>.sh + <id>.json sidecar. Use on /new-agent, 'create an agent', or wanting a reusable button/scheduled agent.

2026-06-070

new-persistent-agent

trevormil/offline-first-sportsbook

Create a global TerMinal persistent memory agent. Use when the user runs /new-persistent-agent or asks for a memory-aware agent that keeps its own INSTRUCTIONS.md, MEMORY.md, STATE.md, JOURNAL.md, and artifacts directory.

2026-06-070

/test-suite — Run tests, report in chat

Ad-hoc test runs. Detects the runner per .agents/testing.md, runs the suite once, and reports results in chat. Writes no artifact.

Delegate to Codex

codex exec -C "$PWD" "/test-suite"

Run directly in Claude only if codex is unavailable or the user asks.

Process

Detect the runner per .agents/testing.md (package.json scripts.test → bun test/vitest/jest; pyproject → pytest; Cargo.toml → cargo test; go.mod → go test ./...; etc.). If multiple, prefer the one CI runs.

Install deps if needed (only if unresolved — no node_modules/.venv). Prefer bun install per global §5.

Run once — capture stdout+stderr, exit code, wall time. No retries.

Report in chat:

**Runner:** <bun test / pytest / ...> **Command:** <exact command> **Status:** pass | fail | partial | error **Counts:** N passed, N failed, N skipped (total N) **Duration:** X.Xs <one-line summary; on failure, list test name + key error excerpt>

What NOT to do

Don't write artifact files. That's /code-review's job (the combined .reviews/ artifact).

Don't edit test files to make them pass. Failing tests become findings.

Don't mark pass when tests were skipped due to install failures — that's error.

Don't re-run until green. One run per invocation; flaky tests are findings worth a /ticket.

Activity

After the run, emit a feed event with the result:

# pass: .claude/bin/activity tests-pass "Tests pass · <N> passed" "<runner> in <time>" # fail: .claude/bin/activity tests-fail "Tests fail · <F> failed / <N>" "<runner>"