rossoctl/context-guru/remove-mcp-and-skills

Run this from this skill directory and print its output exactly as it is in one code block:

python3 ../../scripts/cg.py remove-mcp-and-skills

Append details when the user asks for more. With tool deferral (the ultrahigh preset), the save-more skill's tools view is the newer per-tool view; this one stays until its owner agrees to remove it. In the output, read /context-guru:remove-mcp-and-skills as this skill.

If the output has no Question: line, stop. Otherwise, after printing it, ask exactly that question (yes/no) and wait. Only on yes, run python3 ../../scripts/cg.py followed by every word after /context-guru:remove-mcp-and-skills on the Apply it? line, unchanged, with sandbox_permissions="require_escalated" (it changes the proxy's configuration), and print its output. On anything else say "Nothing changed." and stop. If the user typed apply tools <token> or revert tools themselves, that is their yes: run python3 ../../scripts/cg.py apply tools <token> (or revert tools) the same way. To undo, run the Undo: line the same way.

If loopback access is blocked, retry the read-only command with sandbox_permissions="require_escalated". If it still cannot run, say proxy state is unverified rather than guessing.