Back to skill
Skillv1.0.1

ClawScan security

Session Rotate 80 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

Scanner verdict

BenignMar 5, 2026, 2:16 PM
Verdict
benign
Confidence
high
Model
gpt-5-mini
Summary
The skill's code, instructions, and requirements are consistent with its stated purpose of emitting a new-session trigger at 80% context usage; it does not request credentials, network access, or unusual installs.
Guidance
This skill appears coherent and low-risk: it simply computes used/max ratio and prints standardized trigger lines. Before installing, confirm how your agent supplies used_tokens and max_tokens (ensure those values come from the agent runtime and not untrusted input). Note that the script has no cooldown — repeated checks at or above threshold will repeatedly emit triggers, so integrate it into your heartbeat logic (or add a cooldown/state) to avoid spamming new-session events. If you require stricter guarantees, review/modify the script to add cooldowns, logging, or confidence checks appropriate to your platform.

Review Dimensions

Purpose & Capability
okName/description match the included script and SKILL.md. The single Python script and instructions are exactly what you'd expect for a threshold-based session-rotation helper; no unrelated binaries, env vars, or services are requested.
Instruction Scope
okSKILL.md instructs the agent to read current context usage and call the bundled script with used/max tokens. The instructions do not ask the agent to read unrelated files, access credentials, or send data externally. They are specific and scoped to the purpose.
Install Mechanism
okNo install spec; the skill is instruction-only with a small included script. No downloads, package installs, or archive extraction are performed.
Credentials
okThe skill requests no environment variables, credentials, or config paths. The functionality (token ratio check → printed triggers) does not require secrets or external service access.
Persistence & Privilege
okalways is false and model invocation is allowed (default). The skill does not modify other skills or system settings and has no persistence or privileged installation steps.