Skillv0.0.2

ClawScan security

Willhaben Search · ClawHub's context-aware review of the artifact, metadata, and declared behavior.

BenignMar 10, 2026, 5:16 AM

Verdict: benign
Confidence: high
Model: gpt-5-mini
Summary: This instruction-only skill appears internally consistent: it documents and uses a public search API (api.nochda.at) for willhaben listings, requires no credentials or installs, and its instructions stay on-topic.
Guidance: This skill is instruction-only and appears coherent, but before installing: (1) confirm you trust the third-party API host (api.nochda.at) and its privacy/usage terms, (2) verify linking to willhaben listings is acceptable under willhaben's terms of service, (3) be aware that clicking listing URLs will take you to willhaben pages (phishing risk if domains look similar), and (4) monitor API rate limits if you expect heavy use. If you need an officially supported integration, ask whether this is an official API or a third-party aggregator.

Purpose & Capability: noteThe name/description, endpoints, and declared behaviors align: it's a search/browse/listing-details wrapper for willhaben listings via api.nochda.at. Minor note: the homepage/API domain (api.nochda.at) is a third-party domain rather than the official willhaben domain — that is plausible for an aggregator but worth verifying if you expect an official integration.
Instruction Scope: okSKILL.md only instructs the agent to call documented HTTP GET endpoints on the listed base URL, handle JSON responses, and respect rate limits. It does not ask the agent to read local files, access unrelated env vars, or exfiltrate data to unexpected endpoints.
Install Mechanism: okThere is no install spec and no code files; nothing is written to disk or installed on the agent host. This is the lowest-risk installation model.
Credentials: okThe skill declares no required environment variables or credentials, and the instructions claim the API is public/no-auth. There are no disproportionate secret requests.
Persistence & Privilege: okThe skill is not always-enabled and does not request persistent privileges or modify other skills/config. Autonomous invocation is allowed (platform default) but not combined with other concerning permissions.