HashimTheArab/agent-workflows

codex-review

Use when running Codex CLI code review for uncommitted changes, a PR branch, or a specific commit, especially for second-model review or merge-readiness closeout.

macos-resource-triage

Diagnose what's overloading a hot or sluggish Mac. Checks load average against core count, ranks top CPU/memory consumers, and walks process trees to find runaway agent sessions (leftover AI coding-assistant sessions are a common cause). Use when a Mac's fans…

design-product-identity

Use when creating, replacing, refining, or shipping a product logo, brand mark, wordmark, favicon, application icon, or identity system across a real product.

adversarial-qa

Use only when the user explicitly requests adversarial, hostile, end-to-end, or release-gate QA beyond the project's normal verification suite.

best-practice-research

Use when a technical decision depends on current official documentation, upstream source, standards, release notes, or comparative best-practice evidence.

clarify-requirements

Use when a requested change is genuinely ambiguous about its goal, boundaries, acceptance criteria, or irreversible tradeoffs; skip when repository context supports a safe, reasonable interpretation.

code-review

Use when the user requests code review, an explicit execution workflow requires independent review, or a substantial change needs a merge-readiness assessment.

deep-analysis

Use when the user asks to analyze, investigate, explain why something happens, or provide a grounded cross-file explanation without modifying code.

deslop

Use only when the user explicitly asks to remove AI slop, simplify generated code, clean up unnecessary abstractions, or perform a behavior-preserving maintainability pass.

dispatching-parallel-agents

Use when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies

finishing-a-development-branch

Use when implementation is complete, all tests pass, and you need to decide how to integrate the work - guides completion of development work by presenting structured options for merge, PR, or cleanup

receiving-code-review

Use when receiving code review feedback, before implementing suggestions, especially if feedback seems unclear or technically questionable - requires technical rigor and verification, not performative agreement or blind implementation

subagent-driven-development

Use only when the user explicitly requests agent-driven execution of an accepted plan whose tasks can be delegated independently in the current session.

systematic-debugging

Use when encountering a bug, unexpected test failure, or unexplained behavior before proposing fixes; do not use for an intentional TDD RED failure that fails for the expected missing behavior.

test-driven-development

Use when implementing testable behavior or fixing a bug, before writing production code; skip documentation, metadata-only changes, generated artifacts, and configuration with no executable behavior.

using-git-worktrees

Use only when the user explicitly requests an isolated worktree or a selected execution workflow requires isolation from the current workspace.

verification-before-completion

Use when about to claim work is complete, fixed, or passing, before committing or creating PRs - requires running verification commands and confirming output before making any success claims; evidence before assertions always

writing-plans

Use only when the user explicitly requests a written implementation plan or selects a plan-driven execution workflow for substantial multi-step work.

writing-skills

Use when creating new skills, editing existing skills, or verifying skills work before deployment