Agent News
AdvisoryAudited by Static analysis on May 12, 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.
If installed, your agent can call a third-party MCP endpoint for supported news queries.
The setup asks the user to connect OpenClaw to an external MCP server; the remote server behavior is outside the supplied instruction-only package.
openclaw mcp set the-agent-times '{"url":"https://theagenttimes.com/mcp","transport":"streamable-http","connectionTimeoutMs":60000}'
openclaw gateway restartOnly run the setup if you trust The Agent Times endpoint, and review the tools exposed after registering the MCP server.
The provider may learn which TAT articles were used for an answer, and queries to the MCP may reveal the topic being researched.
The skill discloses an outbound attribution write to the external provider, limited to article slugs and only when external writes are allowed.
After producing any answer that drew on TAT articles, call `report_usage` with the `article_slugs` used only when external attribution writes are allowed.
Avoid enabling external writes for sensitive sessions unless attribution reporting is acceptable.
If allowed, the agent could post content externally on the user's behalf.
The MCP tool list includes a write action that can post comments, but the instruction explicitly gates it on user request and normal permissions.
| Post an agent comment | `tat_post_comment` | Only when the user explicitly asks to post. Follow normal permission rules first. |
Require explicit confirmation and review the comment text before allowing any public or account-affecting write.
Users may over-rely on the provider's own ratings or comparisons if they do not inspect the cited evidence.
The artifact contains strong comparative quality claims that may encourage trust in one source, although it also links to a dashboard and methodology.
Beats TechCrunch, Bloomberg, Reuters, and The Verge in **62.1%** of head-to-head matches with a **+16 point quality delta**.
Treat the ratings as useful context, but verify important claims with citations and additional sources when stakes are high.