Content Research

AdvisoryAudited by Static analysis on Apr 30, 2026.

Overview

No suspicious patterns detected.

Findings (0)

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.

What this means

Your research topic and selected URLs may be sent to search/fetch services, and fetched web pages may influence the generated summaries and drafts.

Why it was flagged

The skill explicitly directs the agent to use external search and fetch tools. This is central to its research purpose and bounded by topic/recent-news workflow, but it is still tool-mediated web access.

Skill content
web_search(query="[topic] news", freshness="pw") ... web_fetch(url="[URL]", maxChars=8000)
Recommendation

Use it for topics you are comfortable researching through external tools, and review generated summaries against the cited sources before relying on them.

What this means

If you configure an API key or browser-based search, the skill may depend on that external service or browser environment for research.

Why it was flagged

The README describes optional browser automation or an API key for search, while registry requirements list no required credentials. This appears purpose-aligned and optional, not hidden, but it introduces potential account/API access if the user enables it.

Skill content
- Browser automation OR Brave Search API key (optional)
Recommendation

Use a dedicated, least-privilege search API key where possible, and avoid granting broader browser/session access than needed for simple web research.

What this means

Brand guidance placed in the file can shape future drafts and may be reflected in generated content.

Why it was flagged

The skill can reuse a local brand configuration file as persistent context for future content generation. This is disclosed and purpose-aligned, but persistent context can influence outputs if it contains sensitive details or is edited by someone untrusted.

Skill content
For branded content, create a `brand-config.md` file with your voice guidelines ... When generating branded content, reference your brand config for consistency.
Recommendation

Keep brand-config.md limited to non-secret style guidance, review it periodically, and do not include passwords, private strategy, or sensitive account details.