Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add RasaHQ/rasa-agent-skills --skill rasa-writing-custom-actionsgit clone --depth 1 https://github.com/RasaHQ/rasa-agent-skillsWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/rasahq/rasa-agent-skills/rasa-writing-custom-actions)<a href="https://agentmods.dev/skills/rasahq/rasa-agent-skills/rasa-writing-custom-actions"><img src="https://agentmods.dev/badge/skills/rasahq/rasa-agent-skills/rasa-writing-custom-actions/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/rasahq/rasa-agent-skills/rasa-writing-custom-actions"><img src="https://agentmods.dev/badge/skills/rasahq/rasa-agent-skills/rasa-writing-custom-actions.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00049 | $0.02477 |
| Opus 5 | $0.00024 | $0.01239 |
| Sonnet 5 | $0.00010 | $0.00495 |
| Haiku 4.5 | $0.00005 | $0.00248 |
Grade A, and why
rasa-writing-custom-actions scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 12d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 269 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Writing Custom Actions in Rasa
Custom actions run Python code when a flow reaches an action step — for example,
calling an external API, querying a database, or computing a value. They execute on an
action server that communicates with the Rasa server via HTTP or gRPC.
Key principle: keep business logic in flows, not in actions. Flows define how the
conversation proceeds (branching, conditions, collect steps). Actions do the raw work
(fetch data, call APIs, query databases) and return results via SlotSet events for
flows to use. Keep each action focused on one responsibility — action_search_products
searches, action_place_order places an order.
If all you need is a simple API call with slot-based input/output mappings and the API
is available on an MCP server, consider using an MCP tool call instead — see
rasa-calling-mcp-tools-from-flows.
Workflow
- Identify what raw work the flow needs — API call, database query, computation.
If the task can be handled with
set_slotsor flow logic alone, don't create an action. - Review existing actions, slots, and responses — reuse before creating new ones.
- Create or open the Python file under
actions/(see "File organization"). - Name the action with a verb after
action_(see "Naming conventions"). - Define the action class:
name()returns the action name,run()does the work (see "Action class structure"). - Read slot values from
tracker.get_slot(), call external services, handle errors, then returnSlotSetevents for the flow to branch on (see "Working with the tracker", "Error handling", and "Returning events"). - Register the action in the domain file under
actions:(see "Domain registration"). - Validate the project.
File organization
Before creating new files, check the existing project layout. If the project already uses a different convention, follow that convention instead.
By default, the action server looks for actions in a file called actions.py or a
package directory called actions/. Use --actions my_module when running the action
server to specify a different module.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 12d ago First seen · 269 lines · 49 tokens per session scan A 9132134e627a
rasa-writing-custom-actions is a skill published in the GitHub repository RasaHQ/rasa-agent-skills (5 stars, last pushed 1mo ago), licensed Apache-2.0. It adds 49 tokens to every session and 2,477 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other skills, from other repositories
stripe-projects
Use after E2B sandbox/API access has been provisioned through Stripe Projects and the user needs to use the resulting E2B API key with the E2B CLI, JavaScript SDK, Python SDK, or Code Interpreter SDK.
azure-mgmt-botservice-py
Azure Bot Service Management SDK for Python. Use for creating, managing, and configuring Azure Bot Service resources. Triggers: "azure-mgmt-botservice", "AzureBotService", "bot management", "conversational AI", "bot channels".
azure-messaging-webpubsubservice-py
Azure Web PubSub Service SDK for Python. Use for real-time messaging, WebSocket connections, and pub/sub patterns. Triggers: "azure-messaging-webpubsubservice", "WebPubSubServiceClient", "real-time", "WebSocket", "pub/sub".
backend
Python server code, APIs, async, strict typing.
fastapi-app
Bootstrap a new FastAPI backend with async SQLAlchemy 2.0, asyncpg, Alembic, Pydantic v2, and no deprecated APIs. Use when the user wants to start, scaffold, or set up a new FastAPI service, a Python REST API, an async backend, or asks to "create a new fastapi app" or "new python backend". Handles JWT auth, layered…
azure-appconfiguration-py
Centralized configuration management with feature flags and dynamic settings.