Adaptive Reasoning
PassAudited by ClawScan on May 5, 2026.
Overview
This instruction-only skill is coherent and low-risk, but it will automatically change the agent’s reasoning mode for complex requests without asking each time.
This appears safe for users who want automatic reasoning-depth adjustment. Be aware it may silently enable deeper reasoning for complex tasks, which can make responses slower or more token-intensive, and the package metadata is slightly inconsistent with the registry entry.
Findings (2)
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.
The agent may spend more time and tokens on complex requests without prompting for confirmation each time.
The skill tells the agent to change reasoning mode automatically and silently. This is purpose-aligned, but users should know it can affect response latency, token use, and session behavior.
**Do not ask. Just activate.** ... Use `session_status` tool or `/reasoning on` command internally before responding
Install this only if you want automatic reasoning-mode changes; use `/reasoning off` or avoid the skill if you prefer explicit control.
It may be harder to confirm exactly which publisher/version this package metadata refers to.
The included package metadata differs from the registry metadata shown for owner, slug, and version. Because this is instruction-only and contains no runnable code, this is a provenance note rather than a security concern.
"ownerId": "kn7azhhp3pdjn1nntp56hm0hj1800etw", "slug": "adaptive-reasoning", "version": "1.0.0"
If publisher provenance matters to you, verify the registry listing and package metadata before relying on it.