
Create Runner Adapter
Guides an agent through adding a new harness/provider integration to the native Paperclip Runner
- Repository
- paperclipai/paperclip
- Stars
- 110,078
- Licence
- MIT
- Files
- 1 file
What it does
This skill gives an AI agent a multi-step procedure for adding a new AI provider or coding-tool integration (called an adapter) to Paperclip's new native Runner system. It tells the agent to define a finite release, investigate the target harness, prove one complete end-to-end slice with real tasks, handle events, approvals, recovery and cancellation, verify packaged delivery, and hand off documented results. It points to repository playbook and template documents for the detailed rules and tests rather than containing them itself.
Great for
- Maintainers of the Paperclip repository adding a new Runner harness or provider
- Developers who need a structured checklist for native Runner adapter acceptance testing
Skip it if
- Editing or migrating legacy direct adapters, which the skill explicitly excludes; this is out of scope and could disturb existing agents or records
- People outside the Paperclip codebase, since it depends on files such as doc/RUNNER-INTEGRATION-PLAYBOOK.md, packages/paperclip-runner and tests/runner-e2e that must exist in the checkout
Try asking
- “Add a new provider harness to the native Paperclip Runner following the integration playbook”
- “Investigate what a new coding harness supports and list its capabilities across the Runner layers”
- “Prepare the acceptance ledger and release contract for a new Runner adapter”
Install in 3 steps
- 1
Download the skill
The ZIP contains the 1 original files, the MIT licence and an attribution note.
create-runner-adapter.zip - 2
Add it to your agent
Claude Code (all projects)
mkdir -p ~/.claude/skills && unzip create-runner-adapter.zip -d ~/.claude/skills/Claude Code / Agent SDK (this project only)
mkdir -p .claude/skills && unzip create-runner-adapter.zip -d .claude/skills/Claude apps (web/desktop): Settings → Capabilities → Skills → Upload skill, and choose the ZIP.
- 3
Ask for the task
No special command needed — the agent reads the skill's description and loads it when your request matches. Try one of the prompts above.
Get better results
- Point the agent at the repository checkout first so it can read the playbook and AGENTS.md before editing
- Ask for one complete vertical slice before attempting the full capability set
- Expect real paid provider calls and budget limits during the acceptance phase; set spending limits up front
What's inside
- SKILL.md
Redistributed unmodified from paperclipai/paperclip under the MIT licence. The assessment was drafted with AI from the skill's files on 2026-10-29. Skills can instruct your agent to run code — read SKILL.md and any scripts before installing.