Runtype
Runtype is a platform for shipping AI products: agents, flows, tools, surfaces, records,
schedules, evals, and product templates. Use this skill as the entry point when the user
is asking about Runtype broadly or needs help connecting an agent to the platform.
This skill is intentionally a router. For implementation work, switch to the narrower
skill that matches the job.
First Move
If the Runtype MCP server is available, use it as the live source of truth before giving
schema, catalog, or creation guidance:
get_build_instructions(task="explain-capabilities") for scoping.get_build_instructions(task="build-product") before product construction.get_build_instructions(task="generate-flow") before flow construction.get_platform_documentation(topic=...) for schemas, surface traits, tool catalogs,
SDK docs, Persona embed docs, dashboard links, and type definitions.
If MCP is not connected, use the CLI-to-MCP golden path instead of presenting several
equivalent setup choices:
- Authenticate the CLI with
runtype auth login so this session can keep working.
- Run
runtype install-mcp. It installs this skill, configures the current harness for
https://api.runtype.com/v1/mcp/protocol, and starts client-owned OAuth when possible.
- Continue the current session with CLI commands. Tell the user to restart or reload the
harness because a running agent may not discover a newly configured MCP connection.
- In the next session, use MCP first and call
get_build_instructions before building.
Do not choose an API key merely to avoid a restart. Headless API-key setup is an exception
for environments where browser OAuth is genuinely impossible; follow https://runtype.com/auth.md
only after confirming that constraint with the user. The full public setup script is at
https://runtype.ai.
Route To Focused Skills
Use this routing table instead of loading every Runtype detail into context:
| User intent |
Use |
| Build, deploy, or validate a product with agents, flows, tools, surfaces, records, secrets, schedules, or evals |
runtype-build-product |
| Inspect or modify a live account, debug failures, read logs/traces, compare evals, manage resources |
runtype-admin |
| Embed or theme a Persona chat widget, build fullscreen assistant layouts, use client tokens, or configure WebMCP/browser-side local tools |
runtype-persona |
| Package a product as a distributable FPO template, handle pending secrets, validate import readiness |
runtype-templates |
| Use the TypeScript/Python SDK, CLI, Marathon, playbooks, sandboxes, or code-first stored/upsert/virtual workflows |
runtype-sdk-marathon |
Mental Model
- Product: the container the user ships.
- Agent: an LLM with a system prompt and tools. Start here for most interactive products.
- Flow: a deterministic pipeline. Use for fixed sequences, indexing, batch work, and hot
paths that should be fast and cheap.
- Tool: a typed callable, from built-ins, Orthogonal APIs, MCP servers, external HTTP,
custom code, local SDK tools, flows, or subagents.
- Surface: where users or machines reach the product, including
chat, api, mcp,
mcp_code, webhook, email, slack, schedule, sms, imessage, discord,
whatsapp, telegram, messaging, a2a, and hosted-page.
- Record: Runtype state and memory. It is not a replacement for the user's business
database.
- Eval: surface-level or capability-level comparison. Prefer surface/product evals when
measuring user experience.
- FPO template: the portable distribution format for a product.
When a product has multiple capabilities on one conversational surface, Runtype can
provision an orchestrator that routes each incoming message to the right capability.
Treat surfaces and capabilities as many-to-many.
Durable References
These local references are fallback context only. Prefer live MCP docs when available.
references/primitives.md for the deeper platform model.references/mcp-tools.md for MCP tool groups and selection.references/surfaces.md for surface traits and prompt implications.references/flow-steps.md for flow step categories and common mistakes.references/persona-widget.md for Persona embed details.references/fpo-templates.md for template packaging.references/working-modes.md for dashboard, MCP, REST, SDK, and on-prem tradeoffs.references/recipes.md for worked product shapes.
Do Not Do
- Do not invent payload shapes. Fetch live docs or validate first.
- Do not let this router grow into platform docs. If more detail is needed, fetch
get_platform_documentation(topic="platform-catalog") or a focused topic, then
read the relevant direct MCP resources when the docs point to them.
- Do not inline secret values. Use
{{secret:KEY}} and pending-secret intake.
- Do not store all customer data in records; keep the source of truth where it belongs.
- Do not use Runtype for generic one-off LLM chat unless the user wants to ship a
repeatable product or operational workflow.
Source: hashgraph-online/awesome-codex-plugins → plugins/runtypelabs/skills/skills/runtype/SKILL.md
1---2name: runtype3description: >- Runtype umbrella skill for onboarding and routing. Use when the user mentions Runtype, asks what Runtype can build, needs MCP/CLI/dashboard setup, or wants to scope an AI product on Runtype. Route hands-on product builds to runtype-build-product, live account ops to runtype-admin, Persona widgets to runtype-persona, FPO packaging to runtype-templates, and SDK/CLI/Marathon work to runtype-sdk-marathon. Do not use for generic LLM chat, unrelated AI frameworks, or non-AI infrastructure unless Runtype is explicitly part of the request.4---5
6
7# Runtype
8
9Runtype is a platform for shipping AI products: agents, flows, tools, surfaces, records,
10schedules, evals, and product templates. Use this skill as the entry point when the user
11is asking about Runtype broadly or needs help connecting an agent to the platform.
12
13This skill is intentionally a router. For implementation work, switch to the narrower
14skill that matches the job.
15
16## First Move
17
18If the Runtype MCP server is available, use it as the live source of truth before giving
19schema, catalog, or creation guidance:
20
21- `get_build_instructions(task="explain-capabilities")` for scoping.
22- `get_build_instructions(task="build-product")` before product construction.
23- `get_build_instructions(task="generate-flow")` before flow construction.
24- `get_platform_documentation(topic=...)` for schemas, surface traits, tool catalogs,
25 SDK docs, Persona embed docs, dashboard links, and type definitions.
26
27If MCP is not connected, use the CLI-to-MCP golden path instead of presenting several
28equivalent setup choices:
29
301. Authenticate the CLI with `runtype auth login` so this session can keep working.
312. Run `runtype install-mcp`. It installs this skill, configures the current harness for
32 `https://api.runtype.com/v1/mcp/protocol`, and starts client-owned OAuth when possible.
333. Continue the current session with CLI commands. Tell the user to restart or reload the
34 harness because a running agent may not discover a newly configured MCP connection.
354. In the next session, use MCP first and call `get_build_instructions` before building.
36
37Do not choose an API key merely to avoid a restart. Headless API-key setup is an exception
38for environments where browser OAuth is genuinely impossible; follow `https://runtype.com/auth.md`
39only after confirming that constraint with the user. The full public setup script is at
40`https://runtype.ai`.
41
42## Route To Focused Skills
43
44Use this routing table instead of loading every Runtype detail into context:
45
46| User intent | Use |
47| ----------------------------------------------------------------------------------------------------------------------------------------- | ----------------------- |
48| Build, deploy, or validate a product with agents, flows, tools, surfaces, records, secrets, schedules, or evals | `runtype-build-product` |
49| Inspect or modify a live account, debug failures, read logs/traces, compare evals, manage resources | `runtype-admin` |
50| Embed or theme a Persona chat widget, build fullscreen assistant layouts, use client tokens, or configure WebMCP/browser-side local tools | `runtype-persona` |
51| Package a product as a distributable FPO template, handle pending secrets, validate import readiness | `runtype-templates` |
52| Use the TypeScript/Python SDK, CLI, Marathon, playbooks, sandboxes, or code-first stored/upsert/virtual workflows | `runtype-sdk-marathon` |
53
54## Mental Model
55
56- Product: the container the user ships.
57- Agent: an LLM with a system prompt and tools. Start here for most interactive products.
58- Flow: a deterministic pipeline. Use for fixed sequences, indexing, batch work, and hot
59 paths that should be fast and cheap.
60- Tool: a typed callable, from built-ins, Orthogonal APIs, MCP servers, external HTTP,
61 custom code, local SDK tools, flows, or subagents.
62- Surface: where users or machines reach the product, including `chat`, `api`, `mcp`,
63 `mcp_code`, `webhook`, `email`, `slack`, `schedule`, `sms`, `imessage`, `discord`,
64 `whatsapp`, `telegram`, `messaging`, `a2a`, and `hosted-page`.
65- Record: Runtype state and memory. It is not a replacement for the user's business
66 database.
67- Eval: surface-level or capability-level comparison. Prefer surface/product evals when
68 measuring user experience.
69- FPO template: the portable distribution format for a product.
70
71When a product has multiple capabilities on one conversational surface, Runtype can
72provision an orchestrator that routes each incoming message to the right capability.
73Treat surfaces and capabilities as many-to-many.
74
75## Durable References
76
77These local references are fallback context only. Prefer live MCP docs when available.
78
79- `references/primitives.md` for the deeper platform model.
80- `references/mcp-tools.md` for MCP tool groups and selection.
81- `references/surfaces.md` for surface traits and prompt implications.
82- `references/flow-steps.md` for flow step categories and common mistakes.
83- `references/persona-widget.md` for Persona embed details.
84- `references/fpo-templates.md` for template packaging.
85- `references/working-modes.md` for dashboard, MCP, REST, SDK, and on-prem tradeoffs.
86- `references/recipes.md` for worked product shapes.
87
88## Do Not Do
89
90- Do not invent payload shapes. Fetch live docs or validate first.
91- Do not let this router grow into platform docs. If more detail is needed, fetch
92 `get_platform_documentation(topic="platform-catalog")` or a focused topic, then
93 read the relevant direct MCP resources when the docs point to them.
94- Do not inline secret values. Use `{{secret:KEY}}` and pending-secret intake.
95- Do not store all customer data in records; keep the source of truth where it belongs.
96- Do not use Runtype for generic one-off LLM chat unless the user wants to ship a
97 repeatable product or operational workflow.
98
99---
100
101**Source:** [`hashgraph-online/awesome-codex-plugins`](https://github.com/hashgraph-online/awesome-codex-plugins) → `plugins/runtypelabs/skills/skills/runtype/SKILL.md`
hashgraph-online