Back to skill
Skillv1.0.0
ClawScan security
星期五 · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignMar 2, 2026, 6:54 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- The skill's requirements and instructions are coherent with a programming agent that uses the opencode CLI; nothing in the package requests unrelated credentials, installs, or capabilities.
- Guidance
- This skill is coherent: it expects the opencode CLI and instructs you how to run it. Before using: ensure opencode is a trusted binary on your PATH; avoid running --auto or background sessions in repositories or directories you don't trust; review ~/.opencode/config.toml and any model/API keys configured there so the agent uses only the credentials you intend; and always review automated changes before committing or pushing them.
Review Dimensions
- Purpose & Capability
- okName/description advertise a coding agent built around the opencode CLI and the only declared requirement is the opencode binary — this is expected and proportionate.
- Instruction Scope
- noteSKILL.md directs the agent to run opencode commands against user project directories, to use modes like --auto/--interactive/--read-only, and references config at ~/.opencode/config.toml. This is appropriate for a coding agent, but --auto and background modes can cause automated file modifications and long-running network activity; follow the provided cautions (don't run in untrusted dirs, review changes).
- Install Mechanism
- okInstruction-only skill with no install spec or bundled code — nothing is written or downloaded by the skill itself, which is the lowest-risk install profile.
- Credentials
- noteThe skill declares no required env vars. SKILL.md references opencode's own config (e.g., ~/.opencode/config.toml) and model selection, which may rely on credentials or API keys configured outside the skill; this is expected but users should ensure those opencode credentials are correct and limited.
- Persistence & Privilege
- okalways:false and no requests to alter other skills or global agent settings. The main risk is intended: modes that let opencode auto-apply changes. The skill itself does not demand elevated or persistent platform privileges.