What happens if Claude encounters a blocker mid-batch?

The skill is strictly instructed to stop execution immediately if it hits a missing dependency, a failed test, or an unclear instruction to prevent errors from compounding.

Does this skill help with final branch cleanup?

After all plan tasks are complete, this skill automatically invokes the 'finishing-a-development-branch' sub-skill to verify the final state and prepare the work for completion.

How does the batch execution process work?

Claude executes a maximum of three tasks from your plan at a time, performing specified verifications for each, then pauses to report progress and wait for your feedback before starting the next batch.

Can I modify the plan once execution has started?

Yes. Between every batch, Claude presents its work for review. At this stage, you can provide feedback or update the plan file, and Claude will return to the review step to incorporate your changes.

Executing Plans

Name: Executing Plans
Author: Kingly-Agency

byKingly-Agency

生産性とワークフロー

Implements complex development plans through systematic batch execution and mandatory human review checkpoints.

概要

The Executing Plans skill provides a structured framework for Claude to process detailed implementation plans safely and accurately. By breaking down large roadmaps into controlled batches—defaulting to three tasks at a time—it ensures that every modification is critically reviewed and verified before proceeding. This approach minimizes errors and scope creep by enforcing mandatory checkpoints for human architect review, allowing developers to maintain high oversight while delegating the systematic implementation of code to the AI.

主な機能

Critical pre-execution plan analysis to identify gaps or logic errors
Automated hand-off to branch finalization skills upon plan completion
Mandatory verification steps for every individual task in the plan
Synchronous reporting loops that require human feedback between batches
Systematic batch execution of tasks to maintain focus and control
0 GitHub stars

ユースケース

Managing large-scale refactoring projects that require frequent verification
Implementing feature checklists where precision and order are critical
Executing multi-step architectural changes defined in a technical spec

概要

主な機能

Critical pre-execution plan analysis to identify gaps or logic errors
Automated hand-off to branch finalization skills upon plan completion
Mandatory verification steps for every individual task in the plan
Synchronous reporting loops that require human feedback between batches
Systematic batch execution of tasks to maintain focus and control
0 GitHub stars

ユースケース

Managing large-scale refactoring projects that require frequent verification
Implementing feature checklists where precision and order are critical
Executing multi-step architectural changes defined in a technical spec