runpod-migrate skill (Runpod): migrate code from GraphQL or REST v1 to REST v2
Runpod's own skill that has a coding agent inventory a repo's Runpod API calls and migrate them from GraphQL or REST v1 to REST v2.
- Installs
- 11.4K
- on skills.sh
- Stars
- 47
- on GitHub
- Files
- 11
- besides SKILL.md
- Updated
- Oct 2026
- last repo commit
runpod-migrate makes your coding agent move a codebase from the Runpod GraphQL API or REST v1 to REST v2. It first runs an inventory scanner over your repo, shows the table of call sites, briefs you on breaking changes, then rewrites one file at a time and checks the result.
It edits your code and writes a runpod-api-inventory.md file into your repo. Verification can create and delete real Runpod resources, so read what it plans before you let it run.
Good for
- Developers whose code calls the Runpod GraphQL API or REST v1
- Teams who want a reviewed migration plan, file by file
- Anyone who wants to know which Runpod API version their code uses
Install runpod-migrate
Claude Code
- 1
Copy this prompt.
Install the runpod-migrate skill in Claude Code. Run this command: /plugin marketplace add runpod/runpod-plugins-official /plugin install runpod@runpod /reload-plugins Then check that it loaded and tell me how to use it.
- 2
Paste it into Claude Code. It shows the change and asks before running it.
Pick one route, not both. Install the whole set, not only this skill: the runpod skills hand off to each other, and the worked golden-path examples live in the runpod router skill's folder. The plugin also registers the hosted Runpod MCP server, which stays inert until you sign in with /mcp.
From the project README. Source.
What's in the skill
When it loads
Migrate a codebase from the Runpod GraphQL API or REST v1 to REST v2 — inventory which parts use which API version, rewrite the call sites, flag breaking changes, and verify. Use when someone asks to move to v2, asks what v2 would change, or asks which Runpod API their code is on. For managing infrastructure rather than migrating code, use runpod-mcp or runpodctl.
Sections in SKILL.md
- Before you touch any code
- The workflow
- 1. Inventory — never migrate what you have not counted
- 2. Brief the breaking changes — before the diff, not after
- 3. Plan, split into required vs cleanup
- 4. Migrate, one file per commit
- 5. Verify against the live API
- 6. Summarize — this is the artifact they will actually read
- Required for the migration
- Cleanup enabled by v2
- Behavior changes to watch
- Still on GraphQL (no v2 equivalent)
- Unlocks: what you can build now
- Ground truth: check the spec yourself
- Precedence when sources disagree
- Tooling notes
Key facts
- Author
- Runpod (vendor's own repo; the marketplace file and plugin manifest name Runpod as owner and author)
- Installs on skills.sh
- 11,448 as of 2026-11-02
- What is inside
- 12 files including SKILL.md: 6 reference guides, 3 eval cases, 1 scanner script (scripts/rp_api_inventory.py) and 1 test file.
- Last change to this skill
- 2026-10-26
What it includes
Counts call sites by API generation
python3 rp_api_inventory.py scans your repo with regex and prints a table: GraphQL, REST v1, already on v2, the unchanged serverless job API and SDK wrappers. It is stdlib-only Python.
Try: “Use runpod-migrate to show me which Runpod API my repo is on.”
Runs code ·
Inventory scannerall, rest or graphql from what you say
The agent states which scope it matched and moves on, without a questionnaire.
Read ·
Scope inferenceTells you which changes apply to your code
Separates loud renames from quiet same-name changes, such as flashboot going from a boolean to a three-value enum.
Read ·
Breaking-change briefRequired, cleanup, and decisions you must make
It stops and asks before touching things v2 removed, such as spot pods or countryCodes.
Read ·
Plan in three bucketsRewrites call sites using the mapping tables
Migrates shared helpers first, never mixes behavior changes with the API change, and can add a RUNPOD_API_V1=1 rollback flag.
Write ·
Migrate one file at a timeRe-scans and exercises real paths
--fail-on-legacy exits 1 if v1 remains. Live checks make reads, and create then delete a small resource to prove the shape.
Runs code ·
Verify against the live APIA five-section report you read instead of the diff
Required changes, cleanup, behavior changes, what stays on GraphQL, and what v2 now lets you build.
Read ·
SummaryPlugin command that runs the full workflow
commands/migrate.md takes a scope and a path, for example /runpod:migrate rest src/.
Runs code ·
/runpod:migrate
From the official docs
Permissions
Medium risk
The agent edits your source files and can create and delete real Runpod resources while verifying. Run it on a clean git branch and review the plan and diff.
What it can do
- Read your data
- Create and change things
- Delete things
- Read-only mode
Watch out for
It edits your code
SKILL.md says to migrate one file per commit and to show the inventory before editing. Work on a branch and review each commit. github.com
Verification writes cost money
SKILL.md says reads are free but writes cost money, and tells the agent to create, assert and delete on the smallest thing. Never let it test against resources you already use. github.com
It leaves an inventory file in your repo
runpod-api-inventory.md lands in the repo root. SKILL.md says to remove it or gitignore it before handing the migration back. github.com
The scanner reads files and prints results
The script imports standard-library modules only and has no network library in its imports. github.com
The plugin route installs all 16 skills and the Runpod MCP config
The marketplace file lists 16 skills for the runpod plugin. It also bundles .mcp.json, which registers the hosted Runpod MCP server at mcp.runpod.io, and a /runpod:migrate command. The git tree shows no hooks file in the plugin folder. github.com
Security advisories
No published security advisories as of Nov 2, 2026.
Technical details
| Talks to | api.runpod.io and rest.runpod.io (public OpenAPI documents and live API checks) |
|---|---|
| Sign-in and keys | Live verification needs your Runpod API key; the scanner reads files only. |
| Scopes |
|
Questions
What does the runpod-migrate skill do?
It has your coding agent find every Runpod GraphQL and REST v1 call in your repo, brief you on breaking changes, rewrite the calls to REST v2 one file at a time, and verify the result.
How do I install runpod-migrate?
Use the install box at the top of this page. The plugin route adds a /runpod:migrate command that runs the workflow. The scanner needs python3, and SKILL.md allows rg for searching.
Does it need an API key?
Only for live verification. The inventory scanner reads your files. Verification calls the live API, so have a Runpod API key ready and use a test account if you can.
Is runpod-migrate safe?
It edits your code and can create and delete small Runpod resources while testing. Use a clean branch, read the plan, and never point verification at resources you rely on.
Does it change the serverless job API?
No. SKILL.md says api.runpod.ai/v2 job calls (run, runsync, status, stream, cancel) are a different API and must not be rewritten. The scanner reports them separately.
What cannot move to v2?
SKILL.md says account and billing identity (myself) and spot or interruptible pods have no v2 equivalent and stay on GraphQL, even in a full migration.
Is this the official Runpod skill?
Yes. It is in Runpod's own runpod-plugins-official repo, and the plugin manifest names Runpod as author.
Does it work in the Claude app or Codex?
The README documents Claude Code and Codex plugins, plus the skills CLI for other agents. The skill runs a Python script and edits files in your repo, so there is no Claude app upload option here.
Information
updated Nov 2, 2026
- Repository
- runpod/runpod-plugins-official
- Folder
plugins/runpod/skills/runpod-migrate
Sources
Docs, issue trackers and discussions this page is built from.
Show all 6Hide
- README.md · github.com
- marketplace.json · github.com
- SKILL.md · github.com
- rp_api_inventory.py · github.com
- migrate.md · github.com
- README.md · github.com