code-forge

Use when you want end-to-end implementation from feature spec or prompt to working code — auto-chains test-cases → plan → impl → review → verify. Works with or without existing documentation.

Use when encountering any bug, test failure, or unexpected behavior — enforces root cause investigation before fixes

Use when implementation is complete and you need to merge, create a PR, or clean up — guides branch integration with 4 structured options

Use when fixing a bug — traces root cause across 4 levels and syncs upstream documents. Supports --repos for parallel multi-repo fixing.

Show all code-forge commands and usage guide

Use when implementing a feature — executes TDD tasks via sub-agents with state tracking and auto-resume. Supports --repos for parallel multi-repo implementation.

Use when facing 2+ independent problems that can be solved concurrently — dispatches one agent per problem domain

Use when creating an implementation plan from docs or requirements — generates TDD tasks with progress tracking

Use when porting a project to another language — batch-generates plans for all features in the target language

Use when reviewing code, handling review feedback, or posting a review to a GitHub PR — 15-dimension quality analysis (with mandatory call-graph pre-analysis), structured feedback response, or automated PR review comment

Use when checking implementation progress — shows feature dashboard with task completion and next steps

Use when implementing any feature or fix outside code-forge workflow — enforces Red-Green-Refactor cycle with mandatory test-first discipline. Supports driven mode (@test-cases.md), auto-analysis mode (@source-file or empty), and standalone mode.

Use before claiming work is done, fixed, or passing — requires running verification commands and confirming output before any success claim

Use when starting feature work that needs workspace isolation — creates git worktree with project setup and baseline tests

Use when encountering any bug, test failure, or unexpected behavior — enforces root cause investigation before fixes. Prevents symptom-fixing, masking bugs, and "just try this" approaches. For code-forge features, use code-forge:fix instead.

Use when implementation is complete and you need to merge, create a PR, or clean up — verifies tests pass, presents 4 structured integration options, executes chosen workflow, and cleans up worktrees. Pairs with code-forge:worktree.

Debug and fix bugs with interactive upstream trace-back — diagnoses root cause level, confirms upstream document updates, and applies TDD fixes. Supports --repos flag for parallel bug fixing across multiple repositories.

Execute pending tasks for a feature — TDD-driven implementation with sub-agent isolation and progress tracking. Use when starting to build, implement, or code a planned feature, resuming partially completed work, or running the next task in a code-forge plan. Supports --repos flag for parallel implementation across multiple repositories.

Use when facing 2+ independent problems that can be solved concurrently — dispatches one sub-agent per problem domain for parallel investigation and fixing. For parallel task execution within a feature, use code-forge:impl instead.

Analyze documentation (or a prompt) and generate an implementation plan with task breakdown, TDD steps, and progress tracking. Use when breaking down a feature, creating tasks from docs or requirements, planning implementation work, or turning a spec into actionable steps.

Port a documentation-driven project to a new target language — initializes project skeleton, analyzes reference implementation, and batch-generates plans for selected features. Use when converting a project to another language, rewriting in a different language, or creating a multi-language SDK from existing feature specs.

Display code-forge feature dashboard with task-level progress from state.json, or show detailed status for a specific feature. Use when checking progress, asking "what's left", viewing task completion, or wanting a bird's-eye view of the project.

Use when implementing any feature or fix outside code-forge workflow — enforces Red-Green-Refactor cycle with mandatory test-first discipline. Supports three modes: (1) Standalone — ad-hoc TDD for quick changes, (2) Auto-Analysis — runs the full spec-forge:test-cases analysis pipeline (project profile, four-layer deep scan, multi-dimensional coverage) then implements all cases via TDD, (3) Driven — reads a test-cases.md document and implements each case via TDD.

Use before claiming work is done, fixed, or passing — requires running verification commands and confirming output before any success claim. Prevents false completion claims, unverified assertions, and "should work" statements.

Use when starting feature work that needs workspace isolation — creates git worktree with automatic project setup and baseline test verification. Ensures worktree directory is git-ignored and safe. Integrates with code-forge:impl and code-forge:finish.