Where this fits

You've just run an explore task—say, gza-1234—that dumped a wall of markdown notes, logs, and partial findings. Now you need to hand that raw material to a planning session or an implementation ticket, but you don't want to re-read every line. This skill takes a completed explore task ID and produces a concise, operator-facing synthesis: what was found, what matters most, what's uncertain, and what to do next. It turns a research artifact into a decision-ready brief.

Why agents benefit

• Synthesizes, not dumps: The agent reads the task output and distills it into a few high-signal paragraphs instead of echoing the whole report.
• Extracts what matters: It identifies the most impactful findings and separates them from noise, so you can act without wading through details.
• Surfaces uncertainty: It explicitly calls out gaps or ambiguous results, preventing premature decisions.
• Suggests concrete next steps: It proposes actionable follow-ups—narrower explorations, implementation tasks, or open questions—tailored to your stated decision context.
• Asks for missing context: If you didn't provide a full prefixed task ID or a decision context, it asks before proceeding, avoiding wasted runs.

Practical scenarios

From research to implementation

You've explored a new API's edge cases. The skill turns that exploration into a prioritized list of implementation tasks, flagging which findings are critical for the first sprint.

Narrowing a broad exploration

Your agent ran a wide-ranging exploration of a codebase. The skill highlights the three areas that need deeper investigation and suggests specific follow-up explore tasks.

Handoff to a planning session

Before a stakeholder meeting, you need a one-page summary of what's known and unknown. The skill produces a crisp brief with recommended next steps, saving you from manual note-crunching.

Add it to your agent workflow

Install the skill with:

npx skills add mhawthorne/gza --skill gza-explore-summarize

Then, in your agent, provide the task ID and optional decision context:

Summarize explore task gza-1234. I need to decide whether to refactor the payment module now or after the next release.

The agent will run uv run gza show gza-1234, inspect the logs, and return a structured synthesis with recommended actions.

Compare before adopting

Before adding this skill, evaluate its quality signals: the GitHub repo (11 stars, single author) is young—check commit activity and open issues. Compare it against alternatives like raw gza show output or a generic summarization prompt. Consider whether your workflow already has a handoff step that this skill would replace or complement.

Why it is worth tracking

This skill fills a real gap: turning exploration output into decisions. Its low star count suggests early adoption, but the author's focus on a narrow, well-defined task is promising. Watch for updates that add support for multiple task formats or richer decision frameworks. If you frequently run explore tasks and need faster handoffs, this is worth evaluating now—and its simplicity makes it easy to swap out if something better appears. For a broader view of similar skills, check the OpenAgentSkill listing to see how it compares with other research-oriented skills.