Back to skill
Skillv1.0.0
ClawScan security
ai-engineer · ClawHub's context-aware review of the artifact, metadata, and declared behavior.
Scanner verdict
BenignApr 29, 2026, 11:44 AM
- Verdict
- benign
- Confidence
- high
- Model
- gpt-5-mini
- Summary
- Instruction-only 'ai-engineer' skill is internally consistent with its description: it provides ML/LLM engineering guidance and code templates, requests no installs or credentials, and does not perform unexpected actions.
- Guidance
- This skill is a safe, instruction-only AI engineering assistant and appears coherent with its stated purpose. Before installing or invoking it: (1) do not provide API keys or secrets to the skill unless you explicitly need runtime integrations; prefer scoped, short-lived credentials if required; (2) treat generated code as a draft—review for security issues, secrets leakage, and correctness before running; (3) if you ask the agent to integrate with services (OpenAI, Anthropic, Pinecone, cloud platforms), confirm where credentials are stored and audit any outbound network activity; (4) monitor agent-autonomous actions and revoke access if unexpected behavior occurs. Confidence in this assessment is high because the skill is instruction-only and requests no unusual access.
Review Dimensions
- Purpose & Capability
- okThe name and description match the SKILL.md content (ML, LLMs, CV, frameworks, MLOps). The skill declares no binaries, installs, or environment access—appropriate for a guidance/template skill.
- Instruction Scope
- noteSKILL.md contains advice, best practices, and a Python code template only. It does not instruct the agent to run commands, read files, or exfiltrate data. It references integrations (OpenAI, Anthropic, Pinecone, cloud ML platforms) but provides no steps that require credentials or external endpoints.
- Install Mechanism
- okNo install spec present. As an instruction-only skill, nothing is written to disk and no third-party packages are automatically pulled in.
- Credentials
- okNo required environment variables, credentials, or config paths are declared. Although the documentation mentions services that normally require API keys, the skill itself does not request or access them.
- Persistence & Privilege
- okalways is false and the skill does not request persistent system privileges or modify other skills/configuration. Normal autonomous invocation is allowed (platform default).