// Apply when implementing features, fixing bugs, debugging errors, investigating failures, tracing root causes, reviewing tech debt, tracing issues, planning fixes, or completing any task. Enforces parallel sub-agent implementation, independent adversarial review, and a 95% confidence gate before stopping.
| name | qa-sweep |
| description | Apply when implementing features, fixing bugs, debugging errors, investigating failures, tracing root causes, reviewing tech debt, tracing issues, planning fixes, or completing any task. Enforces parallel sub-agent implementation, independent adversarial review, and a 95% confidence gate before stopping. |
| effort | high |
Follow this protocol on every implementation, fix, debugging, or review task.
After implementation, spawn a FRESH sub-agent that has not participated in any prior work. Give it this directive verbatim:
"Assume all work done by the implementing agent is incorrect until you can prove otherwise with absolute evidence from the actual code. The implementing agent makes frequent mistakes and tends to miss edge cases. Do not trust any claim without tracing it yourself. Review every change, test, and edge case end-to-end through the real source."
The review agent must:
Spawn a SECOND independent agent to verify original planned work vs. delivered work:
"Assume nothing was completed correctly or fully. Map every originally planned item to actual code changes and verify each one independently. Do not trust the implementing agent's report."
Do NOT stop until ≥95% confident that:
If below 95%, state what remains and continue working.
Enable a high-quality swarm-like Claude Code workflow for the current session, and optionally execute a task immediately using that mode. Uses parallel subagents for breadth, independent reviewer validation for precision, and critic challenge for final confidence. Use when the user wants swarm-like behavior, higher review rigor, or maximum quality without sacrificing Claude Code speed.
End-to-end PR workflow for opencode-swarm. Covers branch setup, conventional commits, mandatory release notes, 5-tier CI checks, and PR submission format. Load this skill before creating branches, commits, or pull requests.
Deep technical debt and CI stability audit for identifying test theater, missing or mis-scoped tests, actual and potential test failures, flaky-test risk, dependency/toolchain brittleness, and structural debt that prevents PRs from going green safely.
Apply when writing tests, modifying test files, fixing test failures, debugging CI failures, adding test coverage, creating adversarial tests, or reviewing any file under tests/. Also apply when implementing features or fixes that require corresponding test changes. Enforces bun:test framework rules, mock isolation, cross-platform compatibility (Linux, macOS, Windows), and CI pipeline awareness. Load this skill before touching any test file.
Apply when committing, pushing, opening a PR, writing a pull request, creating release notes, or updating a changelog. Enforces conventional commit format, mandatory release notes, 5-tier test suite, SHA-pinning for workflow changes, and correct PR body format.
Apply when planning fixes, investigating tech debt, architecting solutions, or diagnosing unknown issues. Search online for current documentation and state-of-the-art approaches before tracing through code.
