Productivity & Workflow Agent Skills

Use when coordinating a repository implementation from a goal through approved, durable, dependency-aware work across one or more agents…

Inspect and map moderate- to high-complexity work as a directed acyclic graph before implementation.

Pressure the facts behind a claim before a decision closes — counter-question with a concrete example, capped attempts, escape hatches…

Turn a finished grill-me (or any design-finalizing) session into a running implementation kickoff.

Grill the user relentlessly about a plan, decision, or idea.

Use when working on a hard problem where most attempts fail — when 99 of 100 things won't work and the temptation is to give up before the…

Delegate coding to xAI Grok Build CLI (features, PRs).

Transcribe audio or video from a local file path or public media URL with Groq, with a read-only dependency preflight that suggests manual…

Check property tax value under BewG §§ 218 ff.: federal model, income value, material value, land value, property type, tax assessment rate…

Decision Support & Consensus Guide. Use when a user needs help navigating a complex goal by understanding domain consensus, key…

Archive a completed change in the experimental workflow.

Continue working on an OpenSpec change by creating the next artifact.

Propose a new change with all artifacts generated in one step.

When to activate: habit design, habit formation, habit loop, implementation intentions, habit stacking, behavior change, atomic habits…

Strategically analyze, select, design, build, review, demo, pitch, and submit differentiated hackathon projects.

Use when the user asks to clean up, refactor, tidy, or productionize a hackathon project or GitHub repository — one that likely has dead…

Generate a consolidated HANDOFF.md at project root that gives a new session (or different CLI tool) full context about what just happened.

Orchestrate free opencode workers from Claude Code to cut token costs.

/em:hard-call — Framework for decisions with no good options.

Agent-only reference for firstmate harness operations.

Transform Claude Code into a fully autonomous agent system with persistent memory, scheduled operations, computer use, and task queuing.