What the reviewer found

Queries three plugin registries (api.claude-plugins.dev, claudepluginhub.com, the Anthropic marketplace) and, only after explicit per-item user approval, writes a fetched community SKILL.md/agent.md to disk; the flagged text is validation logic that warns (not executes) on rm -rf/curl|bash inside a candidate file. No auto-install and no code is run, but installing agent-trusted instructions from third-party repos on stars/verified alone is a real supply-chain surface worth knowing about.

• installs-software — installs software
• network — calls the vendor’s API

What was read

The file as it ships in jikig-ai/soleur:

• .openhands/skills/agent-finder/SKILL.md

What the static scan said

The scan flagged 3things. The reviewer kept 1 and dismissed 2 as false.

• SC2Downloads and executes remote code — false positive
• RMRecursive force delete — false positive
• NETMakes network calls — real

How this review was made

Sonnet 5 read the files above on 7 September 2026 and answered three questions: is it dangerous to whoever installs it, is each scanner finding real, and what should the installer know. The verdict is bound to the file's hash; when the file changes, it is scanned afresh and reviewed again. A script that changes while the definition does not is not re-reviewed — that is a known gap. How the scan and the review work.

Produced by agentmods.dev · Sonnet 5 · 7 September 2026