Skillv1.0.0

ClawScan security

Personal Budget Planner · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignApr 19, 2026, 2:23 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: The skill's code, instructions, and metadata are consistent with a descriptive personal budgeting helper and do not request credentials, make network calls, or perform actions outside the stated scope.
Guidance: This skill appears internally consistent and descriptive only, but exercise normal caution: do not paste full account numbers, passwords, or highly sensitive documents into any skill prompt. If you will enable autonomous invocation or run the code in an environment where outputs are logged, consider reviewing the handler.py locally or running the included tests in a sandboxed environment first. If you plan to deploy this in a shared/managed environment, confirm the skill source/owner is trusted. Overall the skill matches its stated purpose and requests no credentials or network access.

Purpose & Capability: okThe name/description promise templates, analysis, and recommendations. The included handler.py implements only local text parsing and generates JSON recommendations; there are no calls to cloud services or unrelated binaries. Required env vars/configs are none.
Instruction Scope: okSKILL.md only instructs the agent to accept user financial context and produce descriptive outputs. There are no instructions to read system files, network endpoints, or credentials. The handler code processes user text and returns JSON, matching the documented workflow and safety section.
Install Mechanism: okNo install spec provided (instruction-only with a small handler script). No downloads, package installs, or archive extraction. Risk from installation is minimal.
Credentials: okThe skill declares no required environment variables, no primary credential, and no config paths. The code does not access os.environ beyond tests' module path manipulation; it does not accept or request secrets.
Persistence & Privilege: okalways is false, model invocation is allowed (default). The skill does not modify agent configuration or request persistent installation privileges. It runs as a simple handler returning JSON.