MetalHexx
55 verified skills330 total stars
rad-code-review
Review code, phases, and projects for quality, correctness, and conformance. Supports three modes: task review, phase review, and final review. Each mode runs a conformance-first pass against a per-requirement audit table, followed by a lean quality sweep. Status semantics are scope-aware — task and phase use tiered status (on-track / drift / regression); final uses strict status (met / missing).
development6
foo-test-runner
Test runner conventions for the foo package. Use when planning tests inside packages/foo/. Mandates the `foo:vitest` npm script as the test command and a `__foo__` test-file suffix.
testing6
rad-execute
Continue a project through the orchestration pipeline. Ensures the Orchestrator runs as the primary agent — not as a subagent — so it retains full control of agent sequencing. Use for local, background, or cloud-based execution.
testing6
rad-orchestration
Orchestration system runtime, configuration, and context. All pipeline agents load this skill for system context. The Orchestrator receives pipeline-specific guidance.
testing6
rad-execute-parallel
Set up a parallel git worktree for a project and launch orchestration execution in it. Use when asked to "run in parallel", "create a worktree", "execute in a worktree", or when launching a project in an isolated branch for parallel development. Handles git worktree creation, branch setup, source control initialization, and opening the worktree in VS Code, Copilot CLI, Claude Code, or a terminal.
tools6
rad-plan-audit
Audit the two-doc planning set (Requirements + Master Plan) for codebase accuracy, cross-document cohesion, and explosion-readiness. Two modes: self-review (planner-time) and full audit (reporter subagent; orchestrator dispatches a planner subagent to apply fixes).
development6
rad-repo
Use this skill whenever a task might involve code beyond the current working directory — when you're figuring out where code lives, scoping work that may span multiple repositories, or about to act as if the current repo is the whole system — and whenever the user wants to register, bind, describe, group, or manage repositories and repo-groups. The repo registry is your map of the repos a team works across and how they relate.
development6
rad-create-plans
Consolidated document-creation skill for Requirements and Master Plan documents. Routes to the appropriate per-document-type workflow based on the invoking agent's action.
documentation6
rad-orchestration
Orchestration system runtime, configuration, and context. All pipeline agents load this skill for system context. The Orchestrator receives pipeline-specific guidance.
testing6
rad-configure-system
Configure the orchestration system using a structured questionnaire. Walks through system root, project storage, pipeline limits, gate behavior, and source control settings, then generates orchestration.yml.
devops6
rad-execute-coding-task
Execute a single task end-to-end from a self-contained task-handoff document. Supports `code` (with mandatory RED-GREEN) and `doc` / `config` / `infra` task types, plus handoffs that declare no explicit task type. The handoff is the sole input; no upstream planning docs are read.
development6
rad-plan-quick
Start the planning pipeline in quick mode — same plan/audit/approval ceremony as /rad-plan, but with the quick template, Extra Large task size, and autonomous execution mode hardcoded
development6
rad-source-control
Source control operations — commit code or open a PR. All inputs come from the spawn prompt.
development6
rad-plan-audit
Audit the two-doc planning set (Requirements + Master Plan) for codebase accuracy, cross-document cohesion, and explosion-readiness. Two modes: self-review (planner-time) and full audit (reporter subagent; orchestrator dispatches a planner subagent to apply fixes).
development6
rad-approve-plan
Open the post-approval, pre-execution router — delegates to /rad-execute or /rad-execute-parallel
tools6
rad-run-tests
Run the project test suite and report results. Use when executing tests, running unit tests, integration tests, validating code changes, checking test coverage, or verifying acceptance criteria. Provides structured test execution with pass/fail reporting and error details.
development6
rad-orchestration
This skill can be invoked when the user starts talking about planning or executing a project. Use this skill to run a team of subagents through an automated SDLC process that helps them get work done safely and effectively.
data-ai6
rad-plan
Start the planning pipeline for a new project — produces Requirements + Master Plan
devops6
rad-brainstorm
Brainstorm and refine project goals through collaborative ideation. Use when exploring problem spaces, validating concepts, generating UI mockups, building consensus on what to build, creating the project goals and visuals. Trigger when the user talks about brainstorming, goal-setting, idea generation, or early-stage project definition. You can also trigger this skill if they want to generate a mockup, wireframe, architecture or flow diagram, or beautiful html project summary.
development6
rad-plan
Start the planning pipeline for a new project — produces Requirements + Master Plan
devops6
rad-create-plans
Consolidated document-creation skill for Requirements and Master Plan documents. Routes to the appropriate per-document-type workflow based on the invoking agent's action.
documentation6
ui-status
Report whether the radorch dashboard UI server is running, and its URL.
business6
ui-start
Start the radorch dashboard UI as a detached server and report the URL.
business6
ui-stop
Stop the detached radorch dashboard UI server (SIGTERM).
tools6
rad-brainstorm
Brainstorm and refine project goals through collaborative ideation. Use when exploring problem spaces, validating concepts, building consensus on what to build, or creating the initial project goals document.
development6
rad-ui-start
Start the radorch dashboard UI as a detached server and report the URL.
business6
rad-ui-status
Report whether the radorch dashboard UI server is running, and its URL.
business6
rad-execute-parallel
Set up a parallel git worktree for a project and launch orchestration execution in it. Use when asked to "run in parallel", "create a worktree", "execute in a worktree", or when launching a project in an isolated branch for parallel development. Handles git worktree creation, branch setup, source control initialization, and opening the worktree in VS Code, Copilot CLI, Claude Code, or a terminal.
tools6
rad-execute
Continue a project through the orchestration pipeline. Ensures the Orchestrator runs as the primary agent — not as a subagent — so it retains full control of agent sequencing. Use for local, background, or cloud-based execution.
testing6
rad-log-error
Log pipeline execution errors to a structured per-project error log. Use when the pipeline returns success: false, when an agent produces invalid output, or when manual intervention is needed. Appends numbered entries to an append-only error log file.
devops6
rad-ui-stop
Stop the detached radorch dashboard UI server (SIGTERM).
tools6
rad-execute-coding-task
Execute a single task end-to-end from a self-contained task-handoff document. Supports `code` (with mandatory RED-GREEN) and `doc` / `config` / `infra` task types, plus handoffs that declare no explicit task type. The handoff is the sole input; no upstream planning docs are read.
development6
rad-plan
Start the planning pipeline for a new project — produces Requirements + Master Plan
devops6
rad-brainstorm
Brainstorm and refine project goals through collaborative ideation. Use when exploring problem spaces, validating concepts, building consensus on what to build, or creating the initial project goals document.
development6
rad-run-tests
Run the project test suite and report results. Use when executing tests, running unit tests, integration tests, validating code changes, checking test coverage, or verifying acceptance criteria. Provides structured test execution with pass/fail reporting and error details.
development6
rad-ui-status
Report whether the radorch dashboard UI server is running, and its URL.
business6
rad-execute-coding-task
Execute a single task end-to-end from a self-contained task-handoff document. Supports `code` (with mandatory RED-GREEN) and `doc` / `config` / `infra` task types, plus handoffs that declare no explicit task type. The handoff is the sole input; no upstream planning docs are read.
development6
rad-source-control
Source control operations — commit code or open a PR. All inputs come from the spawn prompt.
development6
rad-approve-plan
Open the post-approval, pre-execution router — delegates to /rad-execute or /rad-execute-parallel
tools6
rad-code-review
Review code, phases, and projects for quality, correctness, and conformance. Supports three modes: task review, phase review, and final review. Each mode runs a conformance-first pass against a per-requirement audit table, followed by a lean quality sweep. Status semantics are scope-aware — task and phase use tiered status (on-track / drift / regression); final uses strict status (met / missing).
development6
rad-create-plans
Consolidated document-creation skill for Requirements and Master Plan documents. Routes to the appropriate per-document-type workflow based on the invoking agent's action.
documentation6
rad-log-error
Log pipeline execution errors to a structured per-project error log. Use when the pipeline returns success: false, when an agent produces invalid output, or when manual intervention is needed. Appends numbered entries to an append-only error log file.
devops6
rad-ui-start
Start the radorch dashboard UI as a detached server and report the URL.
business6
rad-ui-stop
Stop the detached radorch dashboard UI server (SIGTERM).
tools6
rad-log-error
Log pipeline execution errors to a structured per-project error log. Use when the pipeline envelope returns ok: false, when an agent produces invalid output, or when manual intervention is needed. Appends numbered entries to an append-only error log file.
devops6
rad-plan-audit
Audit the two-doc planning set (Requirements + Master Plan) for codebase accuracy, cross-document cohesion, and explosion-readiness. Two modes: self-review (planner-time) and full audit (reporter subagent; orchestrator dispatches a planner subagent to apply fixes).
development6
rad-ui-start
Start the radorch dashboard UI as a detached server and report the URL.
business6
rad-approve-plan
Open the post-approval, pre-execution router — delegates to /rad-execute or /rad-execute-parallel
tools6
rad-code-review
Review code, phases, and projects for quality, correctness, and conformance. Supports three modes: task review, phase review, and final review. Each mode runs a conformance-first pass against a per-requirement audit table, followed by a lean quality sweep. Status semantics are scope-aware — task and phase use tiered status (on-track / drift / regression); final uses strict status (met / missing).
development6
rad-source-control
Source control operations — commit code or open a PR. All inputs come from the spawn prompt.
development6
rad-ui-status
Report whether the radorch dashboard UI server is running, and its URL.
business6
rad-run-tests
Run the project test suite and report results. Use when executing tests, running unit tests, integration tests, validating code changes, checking test coverage, or verifying acceptance criteria. Provides structured test execution with pass/fail reporting and error details.
development6
rad-ui-stop
Stop the detached radorch dashboard UI server (SIGTERM).
tools6
rad-execute-parallel
Set up a parallel git worktree for a project and launch orchestration execution in it. Use when asked to "run in parallel", "create a worktree", "execute in a worktree", or when launching a project in an isolated branch for parallel development. Handles git worktree creation, branch setup, source control initialization, and opening the worktree in VS Code, Copilot CLI, Claude Code, or a terminal.
tools6
rad-execute
Continue a project through the orchestration pipeline. Ensures the Orchestrator runs as the primary agent — not as a subagent — so it retains full control of agent sequencing. Use for local, background, or cloud-based execution.
testing6