Skill flagged — suspicious patterns detected
ClawHub Security flagged this skill as suspicious. Review the scan results before using.
Glancify
v1.0.0A proxy service that wraps external web pages with Glancify, enabling interactive content visualization and keyword extraction.
⭐ 0· 30·0 current·0 all-time
byToolWeb@krishnakumarmahadevan-cmd
MIT-0
Download zip
LicenseMIT-0 · Free to use, modify, and redistribute. No attribution required.
Security Scan
OpenClaw
Suspicious
medium confidencePurpose & Capability
The name/description (a proxy that wraps external pages and extracts keywords) matches the SKILL.md and openapi.json: both describe endpoints that forward/transform arbitrary URLs. That capability legitimately explains contacting an external API. However, the skill's documentation advertises paid plans and an API gateway but does not declare any credentials or auth requirements in the skill manifest — a mild mismatch worth noting (the live service may require keys not captured here).
Instruction Scope
SKILL.md instructs the agent to wrap arbitrary target URLs via the external endpoints (e.g., api.toolweb.in/tools/glancify/view). It gives no guidance or restrictions about sensitive or private URLs (internal IPs, authentication‑protected pages, documents, or files). Because using the skill will transmit target URLs and likely their contents to a third party, this creates a realistic exfiltration/SSRF/privacy risk that is not mitigated or documented.
Install Mechanism
No install spec or code is included; the skill is instruction-only. This lowers the risk of arbitrary code being written/executed locally. The openapi.json is just an API surface description and contains no implementation.
Credentials
The skill declares no required environment variables or credentials, which is proportionate for a read-only documentation-only skill. However, the SKILL.md references paid plans and an API gateway; in real use the external service may require keys or billing. The absence of declared credentials means the skill will either call unauthenticated endpoints or require manual credential provisioning outside the manifest — this is a practical mismatch to be aware of.
Persistence & Privilege
The skill does not request persistent privileges (always:false), does not install or modify system configuration, and does not declare special agent-level presence. Autonomous invocation is allowed by default but is not combined with other high privileges here.
What to consider before installing
This skill forwards arbitrary target URLs and (implicitly) page content to a third-party API (toolweb.in). Before installing or using it: (1) Do not allow the skill to be used on internal, private, or authenticated URLs — it can leak sensitive content. (2) Verify the vendor (toolweb.in) and review their privacy/security policy and TLS endpoint; confirm whether API keys or billing are required. (3) Prefer testing only with non-sensitive public pages. (4) If you plan to enable autonomous invocation, restrict or require explicit user approval before the agent calls the skill. (5) If you need similar functionality but cannot trust the vendor, consider running a local/self‑hosted wrapper so content never leaves your environment.Like a lobster shell, security has layers — review code before you run it.
latestvk9787p8bq9ftzmsk3wa87rmv1s843vve
License
MIT-0
Free to use, modify, and redistribute. No attribution required.