v1.0.0

Openclaw Multi Search Engine

BenignClawScan verdict for this skill. Analyzed May 1, 2026, 7:23 AM.

Analysis

This is an instruction-only search helper with no code or credentials, but users should notice that searches are sent to external providers and that some documentation includes dual-use advanced search examples.

GuidanceThis skill appears safe as an instruction-only search helper, but install it with the understanding that your search terms go to external search engines. Avoid putting secrets or confidential information in queries, use advanced search operators only for authorized purposes, and verify the inconsistent metadata before relying on the package identity.

Findings (4)

Artifact-based informational review of SKILL.md, metadata, install specs, static scan signals, and capability signals. ClawScan does not execute the skill or run runtime probes.

Abnormal behavior control

Checks for instructions or behavior that redirect the agent, misuse tools, execute unexpected code, cascade across systems, exploit user trust, or continue outside the intended task.

Tool Misuse and Exploitation

SeverityLowConfidenceHighStatusNote

references/international-search.md

`inurl:login admin` ... `intext:password filetype:txt`

The reference documentation includes advanced search examples that could be used for sensitive or dual-use discovery, such as locating login pages or exposed password text.

User impactIf used carelessly or autonomously, these search patterns could lead the agent toward inappropriate searches for exposed credentials or administrative pages.

RecommendationTreat the advanced operator examples as reference material only; use them for legitimate, authorized searches and avoid credential-discovery or unauthorized reconnaissance queries.

Agentic Supply Chain Vulnerabilities

SeverityInfoConfidenceHighStatusNote

_meta.json

"slug": "tavily-search", "version": "1.0.0"

This included metadata file names a different skill slug and version than the multi-search-engine artifacts, suggesting stale or inconsistent packaging metadata.

User impactThe mismatch does not show malicious behavior, but it makes the package provenance and version history less clear.

RecommendationBefore installing, verify that the registry listing, SKILL.md, and metadata files refer to the same intended skill and publisher.

Human-Agent Trust Exploitation

SeverityInfoConfidenceMediumStatusNote

SKILL.md

## Privacy Engines

- **DuckDuckGo**: No tracking
- **Startpage**: Google results + privacy
- **Qwant**: EU GDPR compliant

The skill presents broad privacy claims about search engines, which may cause users to overestimate privacy protections for submitted queries.

User impactA user might assume sensitive searches are fully private when query data is still being sent to external services.

RecommendationDo not rely solely on the skill's privacy descriptions; review each provider's privacy policy and avoid sensitive searches when privacy matters.

Sensitive data protection

Checks for exposed credentials, poisoned memory or context, unclear communication boundaries, or sensitive data that could leave the user's control.

Insecure Inter-Agent Communication

SeverityLowConfidenceHighStatusNote

config.json

"Google", "url": "https://www.google.com/search?q={keyword}" ... "DuckDuckGo", "url": "https://duckduckgo.com/html/?q={keyword}"

The skill is designed to place user search terms into third-party search-provider URLs, so query contents may be visible to those providers.

User impactAnything typed into a search query, including sensitive names, secrets, or private business context, could be sent to the selected external search engine.

RecommendationUse this skill only for queries you are comfortable sending to the chosen search provider; avoid including passwords, tokens, private documents, or confidential internal details.