implement
User request: $ARGUMENTS
Autonomously execute implementation in-place. Supports plan files, inline tasks, or interactive mode.
Fully autonomous: No pauses except as specified in Phase 3's "Pause ONLY when" list and Edge Cases.
Workflow
Phase 1: Resolve Input & Setup
Review flag: Review workflow runs by default after implementation. If arguments contain --no-review (case-insensitive), disable it. Remove flag from arguments before processing below.
Priority order:
- File path (ends in
.mdor starts with/) → use plan file, optionally with--spec <path> - Inline task (any other text) → create ad-hoc single chunk:
## 1. [Task summary - first 50 chars (if longer, truncate at last space before char 50 if that space is at position 10+; otherwise truncate at char 47 and append "..."; if 50 chars or fewer, use as-is)] - Depends on: - - Tasks: [user's description]
More from doodledood/codex-workflow
research-web
Comprehensive multi-wave web research with strategic source selection. Gathers information from official docs, community resources, and advanced sources. Use for deep technical research, API documentation, best practices. Triggers: research web, deep research, comprehensive research, find documentation.
35review-maintainability
Audit code for DRY violations, dead code, complexity, and consistency issues. Read-only analysis with actionable recommendations. Use before PR or for code quality review. Triggers: review maintainability, code quality, DRY, refactor review.
34review-bugs
Audit code for bugs, logic errors, race conditions, and edge cases. Read-only analysis producing actionable report. Use before PR, after implementation, or when debugging. Triggers: review bugs, find bugs, check for errors, code review.
29review-docs
Audit documentation accuracy against code changes. Identifies stale docs, incorrect examples, and missing documentation. Read-only analysis. Use before PR or after implementation. Triggers: review docs, check documentation, docs up to date.
21explore-codebase
Context-gathering for finding files to read. Maps codebase structure, returns overview + prioritized file list with line ranges. Thoroughness: quick for lookups, medium for bugs/features, thorough for multi-area, very-thorough for architecture audits. Triggers: explore, find files, where is, how does X work.
20review-testability
Audit code for testability design patterns. Identifies business logic entangled with IO, hard-coded dependencies, and mocking friction. Suggests functional core / imperative shell separation. Read-only analysis. Triggers: review testability, testable code, mocking issues, dependency injection.
19