Setup & Installation
Or with the ClawHub CLI, for registry-managed skill folders outside a full OpenClaw workspace:
Version History
Swamp 0.2.0 introduces extension management and swamp-club authentication: - Added step-by-step documentation for authenticating with swamp-club, including login, custom servers, and headless flows. - Introduced new sections on pulling, pushing, listing, and removing Swamp extensions. - Outlined the extension workflow for sharing custom model types via the registry. - Provided guidance on using both models (skills) and extensions for richer sharing and collaboration. - No changes to agent usage or core modeling/test workflows.
What This Skill Does
Swamp is a CLI that models external APIs and services as typed, executable definitions with methods, inputs, and vault-managed credentials. You create models, test them locally, chain them into workflows, and optionally publish them as reusable extensions or *Claw skills.
Typed model definitions with built-in validation and a credential vault prevent the drift and credential leakage common in ad-hoc shell scripts.
When to use it
- Wrapping a REST API into a testable agent capability
- Chaining GitHub and Slack API calls into an automated digest workflow
- Storing API keys in a local vault and referencing them across models
- Publishing a custom model type to a shared team registry
- Converting a validated model into a standalone installable skill
Example Workflow
Here's how your AI assistant might use this skill in practice.
User asks: Fetch open GitHub issues and post a summary to Slack
- 1Run swamp repo init --tool claude to set up the local repository
- 2Create two models: swamp model create command/shell github-issues and swamp model create command/shell slack-webhook
- 3Edit each model definition to configure methods, curl commands, and vault-referenced credentials
- 4Create a workflow with swamp workflow create issue-digest and define job dependency ordering
- 5Run swamp workflow run issue-digest and inspect results with swamp workflow history get issue-digest
Workflow executes fetch and post steps in order, logs each step's output, and delivers a Slack message summarizing the open issues
Requirements
Accounts, API keys, or tools you or your AI assistant may need to set up while using this skill.