Adaptive Reasoning Plus
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 agent has access to files or system tools, it may delete a file too quickly instead of pausing for confirmation or safety checks.
The skill says irreversible actions should force deeper reasoning, but its validation table treats a file-deletion request as a low-complexity direct-execution task. That inconsistency could lead an agent with file tools to act destructively without extra review.
“不可逆性触发时自动升为8+强制推理” ... “Test 2 | "这个文件删了" | 2分 → ⚡直接执行”
Revise the skill so deletion, production changes, financial actions, and other irreversible operations always trigger high-complexity reasoning and explicit user confirmation before execution.
This does not show malicious behavior, but it makes publisher provenance less clear.
The registry metadata lists a different owner ID, “kn776tb7hxs8hrb4h3j08832ad82hebf”, so the package metadata is inconsistent with the registry record.
"ownerId": "kn70e7tf4tkjf8ydnkdnb82dd182ctvt"
Verify the publisher and package source before relying on the skill, especially because the source and homepage are not provided.