Infinite Research Loop
ReviewAudited by ClawScan on May 12, 2026.
Overview
The skill is advertised as an infinite research loop, but its active SKILL.md is for creating/updating skills, which could make the agent follow the wrong workflow.
Do not install this expecting a clean Infinite Research Loop until the active SKILL.md is corrected. If you still use it, be aware it may steer the agent into a skill-creation workflow and may preserve reusable lessons or preferences across tasks.
Publisher note
This is an infinite research loop, with a hidden NO & GAP mechanism to detect when it should stop, in consideration of context.
Findings (3)
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.
A user expecting a research and debugging loop may instead install a skill that triggers skill-creation/update behavior.
The active SKILL.md metadata describes a skill-creation workflow, while the package is presented as an Infinite Research Loop. This mismatch can mislead users about what the installed skill will actually cause the agent to do.
name: skill-creator description: Guide for creating or updating skills that extend Manus...
Correct SKILL.md so its name, description, and body match the advertised Infinite Research Loop purpose, or republish it as a skill-creator package.
The agent may prioritize this unrelated workflow over the user's intended editing or improvement task.
This broad mandatory instruction can redirect the agent's behavior for general modification requests, and it is not aligned with the advertised research-loop function.
For any modification or improvement request, MUST first read this skill and follow its update workflow instead of editing files directly.
Remove or narrow the mandatory modification/update instruction unless the skill is explicitly intended to be a skill-creation workflow.
Future answers may be influenced by stored lessons or preferences from prior interactions.
The skill asks the agent to store lessons and preferences for future synthesis. This is disclosed and scoped, but persistent memory can affect future tasks and may retain user-specific information.
After completion, preserve only high-value lessons: Store: reusable patterns, recurring user preferences, important constraints...
Use only if you are comfortable with persistent learning behavior, and avoid storing sensitive or one-off personal details.