provider-refresh-oct23 — Claude Code skill: how it works and setup
provider-refresh-oct23 configures strict API contracts and routing logic for multimedia generation models. Keeps agent calls aligned with provider updates across speech, image, and video backends.
AGPL-3.0 ★ 78,316 updated 2026-10-24
npx skills add calesthio/OpenMontage --skill provider-refresh-oct23 -a claude-codeWhat provider-refresh-oct23 does
This skill enforces updated endpoint schemas for generative multimedia tools, covering Google, ElevenLabs, HeyGen, Mistral, and xAI. It instructs the assistant to inspect input schemas before invocation, confirm exact hosts and model names, and declare expected generation charges. Automatic fallbacks between providers or models are blocked directly to prevent accidental billing surprises.
The instructions enforce rigid payload boundaries across modalities. Narration requests through eleven_v4 are capped at 2,000 characters, while audio conversion via HeyGen accepts files up to 16 megabytes under a 300-second duration limit. For asynchronous operations on fal or Replicate, the agent must store job metadata and resume polling instead of launching duplicate paid requests.
Direct requests to Google Imagen and Veo will fail unless routed through Vertex backend endpoints.
What it helps with
- route generation jobs across updated multimedia model endpoints
- prevent duplicate paid calls during polling dropouts on external hosts
- validate payload size and character limits before dispatching speech synthesis
- configure dialogue narration with ElevenLabs and HeyGen voice adapters
Example prompts for the agent
- Generate voice narration using
eleven_v4, ensuring context strings stay within limits. - Convert reference speech using
heygen_voice_convertwith a stable idempotency key. - Create an image with Nano Banana 2.1 via Google Vertex and report quoted costs first.
How to install provider-refresh-oct23
1Skills CLI (Vercel)
npx skills add calesthio/OpenMontage --skill provider-refresh-oct23 -a claude-code2Manually: put the files in place
~/.claude/skills/provider-refresh-oct23/SKILL.mdCopy the whole skill folder; for one project use .claude/skills/provider-refresh-oct23/
1Skills CLI (Vercel)
npx skills add calesthio/OpenMontage --skill provider-refresh-oct23 -a codex2Manually: put the files in place
.agents/skills/provider-refresh-oct23/SKILL.mdFor all projects use ~/.agents/skills/
1Skills CLI (Vercel)
npx skills add calesthio/OpenMontage --skill provider-refresh-oct23 -a cursor2Manually: put the files in place
.cursor/skills/provider-refresh-oct23/SKILL.mdCursor also reads .agents/skills/ and .claude/skills/
1GitHub CLI
gh skill install calesthio/OpenMontage .agents/skills/provider-refresh-oct232Manually: put the files in place
.github/skills/provider-refresh-oct23/SKILL.md1Skills CLI (Vercel)
npx skills add calesthio/OpenMontage --skill provider-refresh-oct23 -a gemini-cli2Platform CLI
gemini skills install https://github.com/calesthio/OpenMontage --path .agents/skills/provider-refresh-oct23Manually: put the files in place
.devin/skills/provider-refresh-oct23/SKILL.mdThe legacy .windsurf/skills/ path still works
Claude.ai
Zip the folder and upload it in Settings → Capabilities → Skills
Step by step, with commands for each platform: How to install skills in Claude Code: five ways
Questions about provider-refresh-oct23
How to install provider-refresh-oct23 in Claude Code?
Run npx skills add calesthio/OpenMontage --skill provider-refresh-oct23 -a claude-code in your terminal. The configuration writes to ~/.claude/skills/provider-refresh-oct23/SKILL.md.
What does the provider-refresh-oct23 skill do?
It supplies API contracts and routing instructions for media generation backends. The skill guides the model to validate parameters, respect payload caps, and avoid unapproved model substitutions.
Does provider-refresh-oct23 retry failed jobs on another host?
No, automatic fallback to another host or model is strictly disabled. Pending tasks on platforms like fal or Replicate must be resumed using their saved job paths instead.