Thomashighbaugh/skills/project

Project

Unified entry point for project operations. Each subcommand delegates to an existing command, skill, or agent, with shared behavior for argument parsing and state awareness.

When to Use

• Creating tests for an agent
• Committing, staging, or managing git operations
• Creating or managing pull requests
• Reviewing code quality, security, and architecture
• Analyzing or optimizing code
• Refactoring, simplifying, and modernizing code
• Cleaning up AI-generated code slop
• Generating icon assets
• Organizing files and finding duplicates
• Generating changelogs
• Updating README files
• Tagging releases
• Auditing project health

No-Argument Behavior

When invoked without arguments, list the subcommands as plain text and ask the user to choose. Do NOT call hubMenu or any other tool — just output the list directly. Available operations: create-tests, commit, git-stage-thread, pr, gh, optimize, refactor, simplify, cleanup, modernize, icon, organize, changelog, converge, scan, sandbox, retrospect, purge, release, review, audit, archive, git-cleanup, workspace, readme.

With-Argument Behavior

Directly invoke the matching subcommand. Print the reminder, then delegate.

Subcommand Routing

| Subcommand | Skill/Delegate | What It Does | |------------|----------------|--------------| | create-tests | create-tests command | Generate 8-type test suite for an OpenCode agent | | commit | conventional-commit skill | Create a conventional commit with emoji prefixes | | git-stage-thread | git-stage-thread command | Stage all files modified in the current conversation thread | | pr | github-ops skill | Create, view, merge, manage GitHub pull requests | | gh | github-ops skill | Full GitHub CLI operations — repos, issues, releases, workflows, search | | optimize | inline | Analyze code for performance, security, and maintainability bottlenecks | | refactor | @refactoring agent | Behavior-preserving restructuring — extract, split, reduce coupling | | simplify | @code-simplifier agent | Reduce complexity — flatten nesting, clarify naming, remove abstractions | | cleanup | ai-slop-cleaner skill | Regression-safe deletion-first cleanup of AI-generated code artifacts | | modernize | @refactoring agent | Update code patterns to modern language/framework conventions | | icon | icon-generator skill | Generate web/PWA/UE icon assets from a single source image | | organize | file-organizer skill | Find duplicates, suggest structures, automate file cleanup | | changelog | changelog-generator skill | Generate user-facing changelogs from git commits | | converge | inline | Merge divergent branches and resolve conflicts | | scan | inline | Scan for security vulnerabilities, secrets, and unsafe patterns | | sandbox | inline | Run code in an isolated sandbox environment | | retrospect | inline | Project retrospective — what went well, what didn't, action items | | purge | inline | Remove unused dependencies, dead code, and stale artifacts | | release | inline | Tag and release — bump version, generate changelog, create GitHub release | | review | inline | Full code review round — analyze changes, scan security, check complexity | | audit | inline | Comprehensive project health check — deps, security, quality, coverage, size | | archive | inline | Move stale branches, old artifacts, unused config to timestamped archive | | git-cleanup | inline | Repair CHANGELOG entries with orphaned commit hash references | | workspace | inline | Manage .opencode across projects — list, sync, init, health check | | readme | readme-updater skill | Update README to reflect current codebase state |

Subcommand Behavior

Each subcommand follows the hub pattern:

1. Print a terse reminder (1-2 lines, see Terse Reminders table below)
2. Parse arguments — accept target paths, flags, or operation-specific args
3. Delegate to the appropriate skill, agent, or inline execution (see routing table above)
4. Report results inline with summary of what was done

Terse Reminders

| Subcommand | Reminder on Invoke | |------------|-------------------| | create-tests | Tests: Generate an 8-type test suite for an agent. | | commit | Commit: Analyze staged changes and create a conventional commit with emoji prefixes. | | git-stage-thread | Stage: I'll identify all files modified in our conversation and stage them. | | review | Review: Full code review round — analyze changes, scan security, check complexity. | | pr | PR: Create, view, merge pull requests via GitHub CLI. | | gh | GH: Full GitHub operations via the gh CLI — repos, issues, actions, releases, search. | | optimize | Optimize: Analyze code for bottlenecks, vulnerabilities, and maintainability issues. | | refactor | Refactor: Plan and implement behavior-preserving restructuring. | | simplify | Simplify: Reduce complexity, flatten nesting, clarify naming, remove abstractions. | | cleanup | Cleanup: Remove AI-generated slop — dead code, unused exports, redundant comments. | | modernize | Modernize: Update code patterns to modern language/framework conventions. | | icon | Icon: Generate web/PWA/UE icon assets from a source image. | | organize | Organize: Scan directories, find duplicates, suggest structures, clean up files. | | changelog | Changelog: Generate a user-facing changelog from git commits. | | release | Release: Tag, bump version, generate changelog, create GitHub release in one flow. | | audit | Audit: Comprehensive health check — deps, security, quality, coverage, bundle size. | | readme | Readme: Scan codebase and update README with current state. | | git-cleanup | Git-cleanup: Repair CHANGELOG entries with orphaned commit hash references. |

---

Shared Lifecycle

/project is stateless — no checkpoints or resume. Each subcommand is a one-shot operation:

Step 1: Parse Arguments

Accept target paths, flags, or operation-specific arguments. If no arguments provided, ask the user what to operate on (unless the subcommand is self-targeting like audit or retrospect).

Step 2: Print Reminder

Show the static 1-2 line description for the selected subcommand (see Terse Reminders table above).

Step 3: Execute

Delegate to the appropriate skill, agent, or inline execution per the Subcommand Routing table.

Step 4: Report

Report results inline with a summary of what was done. For destructive operations (purge, archive, git-cleanup), confirm before applying changes.

Related

• /ideation — Plan before you build
• /orchestrate — Execute with a pattern
• /harvest-context — Capture and manage project knowledge
• /init-project — Set up or refresh project configuration