Security audit
Security checks for vulnerabilities and agentic risk
This skill is a purpose-aligned Chinese homestay search tool, but it overstates some capabilities and its AI recommendation command is broken.
Install only if you are comfortable sending lodging search terms, destinations, and dates through the disclosed cloud proxy. Treat the AI recommendation and multi-platform comparison claims as unreliable until the publisher fixes the missing recommendation implementation and documents the proxy token behavior more clearly.
Credentials or environment variables flow to a network sink. This is a high-confidence indicator of credential exfiltration.
method="POST",
)
try:
resp = urllib.request.urlopen(req, timeout=60)
data = json.loads(resp.read().decode("utf-8"))
return data.get("data", data)
except urllib.error.HTTPError as e:
The description broadly matches the travel lodging domain, but several core claims are overstated or unsupported by the supplied code. The implementation only accesses one backend source (Tuniu via a proxy), not multiple tourism platforms. The search capability is present, but keyword/POI-based richness is limited: keyWords and poiName are accepted yet effectively unused in the actual Tuniu query, which always searches for '民宿'. More importantly, the advertised AI recommendation capability is not implemented here because main() calls _tuniu_ai_recommend(args.query), but no such function exists in the code chunk. That makes the recommend action broken rather than functional. Finally, '零配置即装即用' is not accurate because the code relies on an external proxy endpoint and token-based header configuration. These are material description-to-behavior mismatches rather than minor implementation details.
The skill appears to rely on network and possibly environment-backed behavior while declaring no explicit tool scope or permission boundaries. This creates an overprivileged or opaque execution model where users and reviewers cannot clearly see what external access the skill may use, increasing the risk of unintended data egress or misuse of host-provided capabilities.
The skill sends user-supplied travel search data, including destination and travel dates, to an external proxy service without any visible consent, disclosure, or minimization controls. Although this is expected for a travel search integration, undisclosed third-party transmission creates a privacy risk and may violate user expectations or platform policy if sensitive itinerary data is forwarded transparently.
The command descriptions, help text, and user-facing output strings are fixed in Chinese throughout the file, which imposes a specific language on all users. There is no opt-in, locale selection, or documented justification that this skill is intentionally limited to Chinese-speaking users.
The comment and CLI help describe the recommend path as providing AI semantic homestay recommendations, but the implementation invokes _tuniu_ai_recommend(args.query), which is not defined anywhere in the file. This is an active contradiction between the documented intent and the executable behavior, because the feature will fail rather than perform the promised recommendation.
No suspicious patterns detected.