Executing Plans
AdvisoryAudited by Static analysis on Apr 30, 2026.
Overview
No suspicious patterns detected.
Findings (0)
Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.
If the written plan is flawed or contains unsafe instructions, the agent may try to carry them out unless caught during the required review or checkpoints.
The plan file becomes the main driver for the agent’s actions. This is purpose-aligned for a plan-execution skill, and the same instructions require critical review and checkpoints, but users should ensure the plan is trusted and correct.
Follow each step exactly
Use this skill only with plans you have reviewed, and read the batch reports before approving further work.
At the end of the workflow, behavior will depend on the referenced finishing skill as well as this skill.
The workflow explicitly delegates the final development-branch completion step to another skill that is not part of the supplied artifact manifest. This is disclosed and purpose-aligned, but users should ensure that referenced skill is trusted.
REQUIRED SUB-SKILL: Use superpowers:finishing-a-development-branch
Confirm that the referenced finishing-a-development-branch skill is installed from a trusted source and behaves as expected before relying on it.