Usable Stability Loop · The 99/80 Loop
PassAudited by ClawScan on May 4, 2026.
Overview
This is an instruction-only workflow skill that guides the agent toward simple, usable, stable delivery and does not request code execution, credentials, files, network access, or persistence.
This skill appears safe to install as an instruction-only planning aid. Be aware that it will bias the agent toward fast, simple, usable, stable outputs and away from extra polish or speculative architecture.
Findings (1)
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.
When active, the agent may recommend simpler solutions, push back on overbuilding, and stop after a usable stable version rather than continuing to add refinements.
The skill directs the agent's stopping criteria and may cause it to defer extra polish or features. This is disclosed and purpose-aligned rather than deceptive.
Once the main path is usable and stable, explicitly stop unless there is a concrete next pain point.
Use this skill when you want pragmatic delivery; avoid relying on it when you explicitly need exhaustive design, full polish, or comprehensive feature exploration.