tough-tongue/toughtongue-skills / skills/ttai-agent
ttai-agent
VerifiedCreates, edits, and refines Tough Tongue AI Scenarios, the definitions behind voice agents (sales calls, lead qualification, demos, screening, support, booking) and practice roleplays (negotiations, interviews, coaching). Writes ai_instructions, ai_model_config, strategy, tools_config, and rubrik through ttai:create_scenario and ttai:update_scenario. Also resolves the personal or organization workspace, checks plan and balance, places or cancels SIP calls, schedules or cancels meeting bots, mints private Scenario links, grounds a Scenario in the user's documents with knowledge FileBases, and troubleshoots ttai MCP sign-in. Use when the user wants a voice agent or practice Scenario created or improved ("create a voice agent", "the agent ends calls too early", "make it sound more natural"), a call placed, a meeting bot sent, or a Tough Tongue AI account question answered. Not for reports across many sessions (ttai-session-analyst) or browser demo steps (ttai-browser-demo-builder).
references/entities/resources.md
View on GitHub# Resources
Everything in a Tough Tongue AI workspace beyond the account itself
([account-and-access.md](account-and-access.md)). Resources share one permission
and sharing model.
## Contents
- Authoring resources
- Runs and results
- Operational attachments
- Sensitive boundary
## Authoring resources
Each resource: what it is, then what MCP can do with it.
- **Scenario** — one voice (or text) agent definition, versioned. MCP:
`list_resources` / `get_resource` (types `scenarios`, `scenario-versions`);
`ttai:create_scenario` / `ttai:update_scenario`.
- **Collection (course)** — ordered group of Scenarios. MCP: read only (type
`collections`); create and edit in the web app.
- **FileBase** — a folder of files: an Agent Desktop app (kind `app`) or
agentic knowledge (kind `knowledge`). MCP: type `filebases`;
`ttai:create_filebase` / `ttai:write_filebase`. Apps: the `ttai-agent-apps`
skill. Knowledge: attach by ID in `knowledge_filebase_ids`; see
[../mcp/tools.md](../mcp/tools.md#knowledge-filebases).
## Runs and results
- **Session** — one run: transcript, evaluation, extraction, recording. MCP:
type `sessions` (legacy filters for person and date); `ttai:analyze_session`;
`ttai:set_user_metadata` (your own key/value labels).
- **Analytics** — aggregates over sessions: usage, durations, top Scenarios,
members. MCP: `ttai:get_workspace_info(sections: [usage])`.
How runs start: [scenario-engine.md](scenario-engine.md).
## Operational attachments
- **SIP trunk** — customer-owned phone line, inbound or outbound. MCP:
`ttai:list_resources(sip-trunks)`.
- **Bot** — a phone call (`kind: phone`: trunk + Scenario + E.164 numbers) or a
meeting bot (`kind: meeting`: Google Meet, Zoom, or Teams). MCP: type `bots`;
`ttai:schedule_bot`; `ttai:cancel_bot` before it starts.
- **Browser profile** — the browser Tough Tongue AI runs for a Scenario's live
demos. A human signs in to it once through the link from
`ttai:authenticate_browser`; the Scenario stores only the platform-set
`contextId`.
- **Access token (SAT)** — time-limited link to a private Scenario. MCP:
`ttai:create_token` (`scenario_access`), for a runner or for yourself.
- **Custom Function** — HTTP function a Scenario can call. MCP: type
`custom-functions`; attach by ID.
- **Knowledge Base** — legacy RAG: uploaded documents the agent searches with
`knowledge_base_search`. MCP: type `knowledge-bases`; attach by ID in
`knowledge_base_ids`. Not a knowledge FileBase.
- **Voice-agent MCP server** — external tools the agent can call mid-session.
MCP: attach existing IDs via `mcp_server_ids`.
- **Avatar** — system face: static, hybrid, or live. MCP: type `avatar`.
Generic discovery uses `ttai:list_resources` / `ttai:get_resource` with types
such as `avatar`, `custom-functions`, `knowledge-bases`, and `sip-trunks`.
`ttai:read_guide` lists each type's filters.
## Sensitive boundary
Directory types (`sip-trunks`, `knowledge-bases`, `custom-functions`,
`avatar`) return safe metadata plus requested `include_fields`; secret values
come back masked, never document chunks. MCP cannot create Knowledge Bases,
Custom Functions, SIP trunks, or MCP servers. Attach only IDs returned by a trusted workspace call, after loading the
Scenario schema (`ttai:get_schema("scenario")`)
([../scenario/control.md](../scenario/control.md#linked-resources)).
## Key Files
- [account-and-access.md](account-and-access.md) — workspaces and sharing
- [scenario-engine.md](scenario-engine.md) — runtime channels and results
- [../mcp/tools.md](../mcp/tools.md) — tool catalog
Related skills
| Skill | Stars | Updated | Works with | Status | Install |
|---|---|---|---|---|---|
| reverse-engineer-anythingmorluto/reaReverse engineer native, managed, Electron/JavaScript, packaged, firmware, and browser targets with REA, and build agent environments (AgentEnv bundles) of web applications from a live instance or a capture, and of Linux desktop apps with a task from a request. Use shipped-artifact or requested runtime evidence to explain features, compare versions, decompile code, or guide a reconstruction. Skip REA for ordinary source-repository architecture analysis. | Stars: 472k | Updated | Claude CodeCodex | Verified | Install |
| diagnosing-superpowersobra/superpowersUse when a superpowers session went wrong and your human partner wants to know why — repeated work, ignored plans, stumbles, poor results, a skill that didn't fire, "it took too long", "why is it so expensive", "what is it doing" — or wants to build a bug report for the superpowers maintainers, for the current session or a past one identified by id or path, on any harness. | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| dispatching-parallel-agentsobra/superpowersUse when facing 2+ independent tasks that can be worked on without shared state or sequential dependencies | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |
| executing-plansobra/superpowersUse when executing an implementation plan in the current session as the implementer yourself — your human partner chose inline execution, or no subagent tool is available | Stars: 323k | Updated | Claude CodeCodex | Verified | Install |