running-tests
The running-tests skill enables safe execution of focused tests within the opencode-swarm repository using Bun and shell commands. Use this adapter when Codex needs to run isolated per-file tests, perform CI-equivalent validation, check test impact, diagnose failing test output, or execute test loops without interrupting the session.
git clone --depth 1 https://github.com/ZaxbyHub/opencode-swarm /tmp/running-tests && cp -r /tmp/running-tests/.agents/skills/running-tests ~/.claude/skills/running-testsSKILL.md
# Running Tests Use this adapter when executing tests in `opencode-swarm`. For writing tests, load `$writing-tests` instead. Read, in order: 1. `AGENTS.md` 2. `.opencode/skills/running-tests/SKILL.md` Codex-specific execution notes: - Use the available shell execution tool for `bun --smol test` and PowerShell/bash loops. - Prefer `rg` and repo scripts to rediscover exact commands from source. - Capture long output to a temp file when needed and report the important tail or failure lines. - Do not use broad OpenCode `test_runner` scopes for repo validation. - On Windows, if Bun reports `EPERM` after a forced dependency refresh, rerun the same focused command with approved/elevated access before treating it as a test failure. Default rule: one source file or one explicit test file can be targeted narrowly; multiple files or directories should use shell loops or the documented tier commands.
>
Run a rigorous, quote-grounded codebase review or security/QA/accessibility/performance/AI-slop/enhancement audit. Use for full-repo or large-subsystem review reports; not for normal implementation. Performs Phase 0 inventory, selected exhaustive tracks with non-diluting depth, coverage closure, reviewer/critic validation, and writes .swarm/review-v8 artifacts without modifying source files.
>
>
Use when asked to trace, investigate, root-cause, plan, fix, close, or prepare a PR for a GitHub issue or bug report. Runs an evidence-first issue workflow: GitHub intake, reproduction, reasoning-guided localization, no-gap fix planning, independent critic review, user approval gate, implementation, tests, and PR-ready closure.
>
>