v1.0.0

Smart Model Routing for Z.AI

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 5:29 AM.

Analysis

This is an instruction-only model-routing skill that appears aligned with its stated purpose and does not request code execution, credentials, installation, or extra system access.

GuidanceBefore installing, make sure you want the agent to automatically select z.ai/GLM model tiers based on task complexity. There is no code or credential access shown, but for important decisions or sensitive tasks, consider explicitly requesting a stronger model.

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.

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

sessions_spawn(task="Build the REST API endpoint", model="GLM-4.7") ... Always use Flash for cron unless the task genuinely needs reasoning.

The skill gives examples for selecting models when spawning subagents or handling cron-like tasks. This affects agent/model invocation behavior, but it is disclosed and directly related to the stated model-routing purpose.

User impactThe agent may choose cheaper or more capable z.ai models according to the skill’s rules, which can affect answer quality, latency, and API cost.

RecommendationUse this skill if you want automatic model routing, but manually override the model for sensitive, high-stakes, or quality-critical work.