v1.0.0

opus

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 5:27 AM.

Analysis

This is an instruction-only model-routing skill with no code, credentials, or install steps; the main things to notice are model/cost behavior and a minor metadata naming mismatch.

GuidanceThis skill appears safe for users who want automatic Claude model selection. Before installing, verify the package identity because the registry slug and internal slug differ, and make sure you are comfortable with the model escalation rules because they can influence API cost and answer quality.

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.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Tool Misuse and Exploitation

SeverityLowConfidenceHighStatusNote

SKILL.md

Auto-route tasks to the cheapest Claude model that works correctly. Three-tier progression: Haiku → Sonnet → Opus. Classify before responding.

The skill intentionally directs model selection before answering. This is the core purpose and is disclosed, but users should notice that it can change which model is used and therefore affect cost and response quality.

User impactThe agent may choose a cheaper or more expensive Claude model according to the skill's rules, which can affect spending and answer capability.

RecommendationInstall only if you want automatic model routing, and monitor whether the escalation rules match your cost and quality expectations.

Tool Misuse and Exploitation

SeverityInfoConfidenceHighStatusNote

SKILL.md

sessions_spawn(task="Check backup status", model="haiku")

The SKILL.md includes examples for subagent model selection. These appear to be documentation examples rather than automatic execution instructions, but they show the intended use with agent-spawning mechanisms if the environment supports them.

User impactIf used in workflows that spawn subagents, the skill could influence which model those subagents use.

RecommendationTreat the subagent and cron snippets as configuration guidance, and require normal user/workflow approval before creating recurring or delegated agent tasks.

Agentic Supply Chain Vulnerabilities

SeverityInfoConfidenceHighStatusNote

_meta.json

"slug": "smart-model-switching"

The included _meta.json slug differs from the registry slug shown as "opus". This is a packaging/provenance inconsistency, though the skill is instruction-only and no executable code is present.

User impactThe package identity may be slightly confusing because the registry-facing name differs from the internal metadata name.

RecommendationConfirm that the registry entry, package metadata, and publisher are the skill you intended to install.