Frame.io Agent
Enables coding agents to read and write to Frame.io (projects, assets, reviews, comments, and more) with confirmation-gated mutations.
README
Frame.io Agent
by VAXA Studio
Run Frame.io from your coding agent.
Your agent can find your latest cut, summarize the director's notes with timecodes, pull a YouTube reference straight into a project folder, build a contact sheet from your last shoot, and send a curated review link to a client — all without opening Frame.io.
10-minute setup. Works with Claude Code, Cursor, OpenAI Codex CLI, Gemini CLI, or any agent that runs local commands. MCP optional. Read-only by default; every mutation (upload, share, comment) is confirmation-gated.
Unofficial community tool. Not affiliated with, endorsed by, or supported by Adobe or Frame.io. "Frame.io" is a trademark of Adobe.
Quick start (with an agent)
git clone https://github.com/sheldondirector/frameio-agent
cd frameio-agent
Then paste this to your coding agent:
Connect this repo to my Frame.io account, read-only.
1. Read AGENTS.md.
2. Install deps.
3. Set up .env.
4. Run `frameio-agent auth login`.
5. Run `frameio-agent verify`.
6. List my projects.
7. Find my latest cut.
8. Summarize its review comments with timecodes.
Do not print secrets. Prefer --json when parsing.
Quick start (manual)
python scripts/setup.py # install deps
cp .env.example .env # create config
python -m frameio_agent.cli auth login # guided OAuth wizard
python -m frameio_agent.cli verify
python -m frameio_agent.cli projects --json
python -m frameio_agent.cli latest --project <id> --json
python -m frameio_agent.cli comments <file_id> --json
What it does
Read (no confirmation needed):
auth login— guided Adobe IMS OAuth wizard with clipboard auto-detect. First time only.auth status/verify— confirms the connection without printing tokens.projects --json— lists accounts / workspaces / projects.latest --project <id>— newest-updated video assets, recency-first.search "<query>"— account-wide search across all your projects and folders.comments <file_id> --json— normalized review notes (timecode + timestamp_seconds + author + thread).brief --project <id>— one-paragraph project status.
Write (every one is confirmation-gated):
share create <file_id> [<file_id>...] --name "..."— bundle assets into a Frame.io review share. Multi-asset, optional--reviewers email@a.com,email@b.com. Default visibility public;--restrictedfor signed-in only. Prompts y/N before sending;--yesto skip.refs add <url-or-path> --folder <folder_id>— pull a YouTube/X URL via yt-dlp, or upload a local file, or trigger Frame.io's remote_upload for direct URLs Frame.io can fetch. One-stop reference ingestion.contact-sheet --project <id> --out sheet.png— composite thumbnails from a project or folder into a single image grid.
Read-only by default. Mutations only fire after explicit y/N confirmation (or --yes when the user has authorized the action in the current conversation).
Security
- Read-only by default; every mutation is opt-in and confirmation-gated. Share creation, reference uploads, comment posts — all require y/N (or an explicit
--yesthe user authorized). No silent writes ever. - Delete / rename / move / permission-change are still NOT in scope. Even the agent can't accidentally trash your project hierarchy. Those endpoints exist in V4; the CLI just doesn't call them.
- Secrets stay local.
.envand~/.frameio-agent/tokens.jsonare git-ignored. The CLI redacts token-like values from any output. Share passwords are never echoed. - Bring your own credentials. The wizard helps you create an Adobe Developer Console OAuth Web App; nothing is ever sent to the maintainers.
Optional: MCP
If your agent supports MCP, run frameio-agent mcp to expose the same operations as MCP tools. The CLI is the spine — MCP is a thin wrapper.
Troubleshooting
| If you see | Run |
|---|---|
Missing FRAMEIO_CLIENT_ID |
frameio-agent auth login |
Frame.io is not authenticated |
frameio-agent auth login |
Your Frame.io session expired |
frameio-agent auth login |
Found multiple accounts |
Re-run with --account <id> |
Traversal cap reached |
Use latest instead, or pass --folder <id> |
License
MIT. See LICENSE.
Built by VAXA Studio — we make tools for creative teams using coding agents.
Unofficial community tool. Not affiliated with, endorsed by, or supported by Adobe or Frame.io.
Recommended Servers
playwright-mcp
A Model Context Protocol server that enables LLMs to interact with web pages through structured accessibility snapshots without requiring vision models or screenshots.
Magic Component Platform (MCP)
An AI-powered tool that generates modern UI components from natural language descriptions, integrating with popular IDEs to streamline UI development workflow.
Audiense Insights MCP Server
Enables interaction with Audiense Insights accounts via the Model Context Protocol, facilitating the extraction and analysis of marketing insights and audience data including demographics, behavior, and influencer engagement.
VeyraX MCP
Single MCP tool to connect all your favorite tools: Gmail, Calendar and 40 more.
graphlit-mcp-server
The Model Context Protocol (MCP) Server enables integration between MCP clients and the Graphlit service. Ingest anything from Slack to Gmail to podcast feeds, in addition to web crawling, into a Graphlit project - and then retrieve relevant contents from the MCP client.
Kagi MCP Server
An MCP server that integrates Kagi search capabilities with Claude AI, enabling Claude to perform real-time web searches when answering questions that require up-to-date information.
E2B
Using MCP to run code via e2b.
Neon Database
MCP server for interacting with Neon Management API and databases
Exa Search
A Model Context Protocol (MCP) server lets AI assistants like Claude use the Exa AI Search API for web searches. This setup allows AI models to get real-time web information in a safe and controlled way.
Qdrant Server
This repository is an example of how to create a MCP server for Qdrant, a vector search engine.