ML Spec — Testing skill for Claude Code
Full specification loop - plan, write, and review spec (Meta-Loop Steps 2.1-2.4).
How to install ML Spec
Installs to ~/.claude/skills/jpavley-meta-loop-ios-ml-spec/SKILL.md
mkdir -p ~/.claude/skills/jpavley-meta-loop-ios-ml-spec && curl -fsSL https://raw.githubusercontent.com/jpavley/meta-loop-ios/HEAD/commands/ml-spec.md -o ~/.claude/skills/jpavley-meta-loop-ios-ml-spec/SKILL.md Restart Claude Code, or start a new session, for it to be picked up.
What ML Spec does
description: Full specification loop - plan, write, and review spec (Meta-Loop Steps 2.1-2.4) argument-hint: wireframe_path
Meta-Loop: Specification Loop
Complete the specification loop: plan the spec, write it, and review for correctness.
**Modes:** Plan Mode → Agent Mode → Ask Mode
Arguments
$ARGUMENTS: Path to wireframe(s) that will drive the specification- Examples:
specs/views/my-feature/main-view-wireframe.pngorspecs/views/my-feature/
- Examples:
Prerequisites
This
Alternatives in Testing
- Deslop Reborn — One iteration of the IronClaw Reborn de-slop loop — take ONE Reborn crate, fan out parallel review sub-agents 12.6k ★
- Cc10x — The Loop Engine for Claude Code — engineer the loop, not the prompt 163 ★
- Spawn Fs Team — Boot a 5-agent full-stack Flotion team as a new workspace in the cmux window (reuse the open window; only crea 107 ★
Full documentation available on GitHub
View Source RepositoryRelated Skills
Book Write
Draft, review, and finalize one chapter — full pipeline with a pre-draft validity check, the seven-check QA ga
ML Wireframe
Review wireframe to ensure AI understanding before spec creation (Meta-Loop Step 1.2)
Valcraft
Agent skills for spec-driven delivery — one plugin for Claude Code, OpenAI Codex, and OpenCode. At the center:
Feature Team
Run a multi-task plan with the implementer→spec-review→quality-review loop using the model ladder. For COMPLEX
Gsuper Workflow
Run gsuper workflow orchestrator (gsuper-brainstorm → gsuper-write-spec → gsuper-write-plan → gsuper-implement
Happysquad Loop
Run the full happysquad loop (architect → implement → test → conflict-gate → review) on a task, iterating unti
Related Agents
Arquiteto
Meta-persona of the workbench. Invoke to plan complex tasks (>3 steps, multi-domain, irreversible action), cre
Propagate
Generate tests from Allium specifications. Use when the user wants to propagate tests, generate test files fro
Fixme Write Product Spec
Writes product specifications by reading product context and current user-visible behavior. Produces only prod