Productivity & Workflow Agent Skills for ChatGPT

Use when the user invokes /goal, asks for Codex goal mode, or starts long-horizon autonomous work that should be refined into a concrete…

Use only when the user explicitly invokes the legacy $codex-prove command; redirect that request to the canonical $codex-air workflow…

Inspect same-day Codex session activity, score coding momentum across the day, detect idle drop-off, and coach the next small mission to…

Use when starting nontrivial repo work where route choice, scope, assumptions, or completion evidence could drift; defaults to a…

For collegiality and responsibility, organizational duty U: checks deadline, form, responsibility and urgency; result: test product with…

Convert a Notion spec into linked implementation plans, tasks, and ongoing status updates.

Delegate a coding task to the Command Code CLI (`cmd`) as a background implementer, then review its diff and land it yourself.

Render command output (e.g. ls -alr) as a lossless table.

For commentary: Compliance documentation and file note: organizes files, documents, and gaps; result: audit product with risk and next…

Establish and maintain enough shared understanding for ambiguous project work.

Use after benchmark-methodology has produced scored competitor profile cards.

Guides multi-iteration product strategy, reframes ambiguous problems, and transforms exploratory discoveries into reusable cognitive assets.

Connected analyze-plan-execute-learn-repeat cycle with persistent memory.

Garbage collection for your Claude Code configuration.

Minimize token usage through efficient context loading.

The user is back at the keyboard — resume work.

Resume interrupted session. Loads context, companion findings, and git state.

Implements Teresa Torres' continuous discovery habits for weekly customer contact, opportunity solution trees, and assumption testing.

Analyzes contractor engagements against standard classification factors to assess operational exposure and evaluate remediation options.

An addon (best paired with [[work-until-limit]], but works for any long/autonomous run) that gives the user a physical red STOP button — a…

Decompose a claim into facts, conclusions and value judgements, verify every fact online, audit the inference, and say how far to trust it.