Use Tough Tongue AI inside your Claude, Codex & Cursor
We just shipped plugins, MCP, and skills for ToughTongue AI.
What this means: Your coding agent (Claude Code, Codex, Cursor) can now create practice scenarios, refine them from real session data, and analyze team performance - without leaving the terminal.
Setup (one command):
Claude Code: claude plugin install toughtongue@toughtongue-skills
Codex: codex plugin add toughtongue@toughtongue
Cursor: Settings > Plugins > Import from Repo
What ships:
4 skills: scenario-creator, scenario-refiner, session-analyst, and getting-started
26 MCP tools: scenarios, sessions, analytics, SIP calls, meeting bots, collections
What you can actually do with this:
"Pull the last 3 lost deals from Gong and create a practice scenario for the pricing objection."
"For the Enterprise Discovery Call scenario, pull the last 50 sessions. Top 5 improvement areas? Build me a deck."
"The onboarding agent ends calls too early. Pull the low-scoring sessions, diagnose, and fix."
"I have a call with Sarah Chen in 30 minutes. Make me a practice scenario."
Watch the demo:
Get started: https://github.com/tough-tongue/toughtongue-skills
Docs: https://www.toughtongueai.com/agents
What workflows would you build with this?



Replies
i appreciate the MCP tools. how flexible are custom workflows? user defined templates could make adoption much easier for different teams.
Tough Tongue AI
@jason_scott8 Very flexible, though user-defined templates are still an area we should explore more deeply. A lot of our users already create their own custom workflows, skills, and agent-writing patterns on top of our platform. I think the next step is giving them a cleaner way to package and reuse those patterns across teams. I’d also love to explore if anyone has formalized this well elsewhere.
i would love to see more examples from customer support teams instead of only sales. could the same workflow improve onboarding and retention conversations? that seems like a natural expansion.