Agents often lack credentials for live data and external APIs. treg searches priced endpoints and calls approved services through one authenticated interface.
Review the Skillstore skill "treg" from https://skillstore.io/skills/superdesigndev-treg.md and its manifest at https://skillstore.io/api/skills/superdesigndev-treg/manifest. Verify the artifact. Do not auto-install. Inspect the skill and report your findings, then wait for an operator or manual installation decision.
The audit confirms remote pipe-to-shell installation, credential collection, hidden configuration access, and broad authenticated command execution. Many backtick and URL matches are documentation-only false positives, but the confirmed critical and high-risk workflows require strong user consent and verification.
The skill tells agents to request blanket approval for treg and avoid per-command confirmation, even though later commands upload secrets and change accounts.
The instruction explicitly labels additional confirmation as friction and asks for treg approval as a whole.
The proxy relays arbitrary HTTP methods, headers, cookies, and bodies, while treg run executes vendor CLIs locally or remotely with injected credentials.
The documented interface supports state-changing HTTP requests and arbitrary vendor CLI arguments under organizational credentials.
shell commands, request `treg` **once**, as a whole, rather than a prompt per subcommand. Halting
The skill asks for blanket approval of every treg subcommand, reducing oversight for later commands that spend funds, upload secrets, or change accounts.