Riverside + AI - for ChatGPT, Claude, GitHub Copilot, Microsoft 365 Copilot, Gemini, and any agent that speaks MCP
Unofficial. Community-built Claude Code Skill and MCP server for the Riverside
API. Not affiliated with, endorsed by, or sponsored by RiversideFM, Inc..
Full skill page - install, outcomes, safety model.
Riverside API surface as MCP tools. Works with the AI you already use - ChatGPT (Plus/Pro+), Claude Desktop, Codex, Claude Code, Claude Cowork, and GitHub Copilot - plus Microsoft 365 Copilot / Copilot Studio and Google Gemini via the remote path. Free, open source, runs on your laptop. Built for MSP owners. No code required.
Works with your agent
The six agents MSP owners actually use (self-serve, works today):
| Your AI agent | How to install the Riverside skill |
|---|
| Claude Desktop | Run installer, then Settings > Extensions to register riverside-fm-mcp (no JSON editing). |
| ChatGPT (paid plans) | Run installer, expose riverside-fm-mcp over HTTPS, register as a Developer Mode connector. |
| Codex CLI | Paste the install prompt below. |
| Claude Code | Paste the install prompt below. |
| Claude Cowork | Paste the install prompt below. |
| GitHub Copilot (VS Code) | Run installer, add riverside-fm-mcp to mcp.json under the servers key, then pick Agent mode. |
For ChatGPT, the Riverside MCP server is stdio - to use it with ChatGPT you expose it over HTTPS via the supergateway bridge or your own endpoint. See mcp-install.md.
Also for the Microsoft and Google stacks
Big install base, but an honest heads-up: these are the remote / enterprise path, not the local binary you just installed.
| Agent | What it takes |
|---|
| Microsoft 365 Copilot / Copilot Studio | Not self-serve. Host riverside-fm-mcp over HTTPS, then wire it into Copilot Studio (Tools > Add a tool > Model Context Protocol > Server URL) or a declarative agent. Needs a Copilot Studio license + tenant admin. See mcp-install.md. |
| Google Gemini | Gemini CLI is local - same as Claude Code. The Gemini app is remote - same HTTPS path as ChatGPT. See mcp-install.md. |
Skill-native agents (also covered): Hermes and OpenClaw read this skill's SKILL.md directly and speak MCP - see their install sections below. Also works with Cursor, Windsurf, Cline, Continue.dev, and Zed via MCP. Full per-tool wire-up: docs/which-agent.md.
Run more than one agent? Install across all 51+ supported agents in one command: npx skills add Servosity/msp-skills@latest (requires Node.js, then run the per-skill installer for the CLI/MCP binaries). See docs/which-agent.md.
Install in 60 seconds
Fastest for Claude Desktop - one-click .mcpb
Download Riverside MCP (.mcpb) - then open Claude Desktop > Settings > Extensions and select the file. One click, no JSON, no shell. (Browse every Riverside release on the releases page.)
Prefer the Claude Code plugin? Add the marketplace once, then install - works immediately, no directory listing required:
/plugin marketplace add Servosity/msp-skills
/plugin install riverside-fm@msp-skills
Path A - paste one prompt into your AI agent (recommended)
Copy this into Claude Code, Codex CLI, or Claude Cowork:
Install the Riverside Skill and MCP server from Servosity/msp-skills in this agent workspace. If this workspace uses a POSIX shell (macOS, Linux, WSL, or Bash), run bash <(curl -fsSL https://raw.githubusercontent.com/Servosity/msp-skills/main/skills/riverside-fm/install.sh). If it uses Windows PowerShell, run iwr -useb https://raw.githubusercontent.com/Servosity/msp-skills/main/skills/riverside-fm/install.ps1 | iex. Then authenticate per the README and run riverside-fm-cli --help to explore.
The same prompt works in any agent that can run shell.
Path B - run the installer yourself
Windows (PowerShell):
iwr -useb https://raw.githubusercontent.com/Servosity/msp-skills/main/skills/riverside-fm/install.ps1 | iex
macOS / Linux:
bash <(curl -fsSL https://raw.githubusercontent.com/Servosity/msp-skills/main/skills/riverside-fm/install.sh)
The installer drops both riverside-fm-cli (the CLI) and riverside-fm-mcp (the MCP server) into your user bin path. Claude Code, Codex, and Cowork discover the Skill via SKILL.md in this directory.
Verify:
riverside-fm-cli --version
Upgrade to the latest version
The installer always fetches the current release - re-run it to upgrade:
macOS / Linux:
bash <(curl -fsSL https://raw.githubusercontent.com/Servosity/msp-skills/main/skills/riverside-fm/install.sh)
Windows (PowerShell):
iwr -useb https://raw.githubusercontent.com/Servosity/msp-skills/main/skills/riverside-fm/install.ps1 | iex
Claude Desktop .mcpb users: download the latest .mcpb (top of this section) and re-select it in Settings > Extensions. Claude Code plugin users: /plugin update riverside-fm@msp-skills.
Add to Claude Desktop, GitHub Copilot, Gemini CLI, Microsoft 365 Copilot, or another MCP client
After the installer runs, see mcp-install.md and docs/which-agent.md for the per-agent wire-up - one section per agent, including the GitHub Copilot servers key and the remote Microsoft 365 Copilot / Copilot Studio path. Claude Desktop's Settings > Extensions panel is the simplest path; the MCP config block (for users who prefer editing JSON) is documented in mcp-install.md.
Install for Hermes
From the Hermes CLI:
hermes skills install servosity/msp-skills/skills/riverside-fm --force
Inside a Hermes chat session:
/skills install servosity/msp-skills/skills/riverside-fm --force
Hermes speaks MCP natively, so it can also use the riverside-fm-mcp server directly - same install path, same env vars.
Install for OpenClaw
Tell your OpenClaw agent (copy this):
Install the riverside-fm skill from https://github.com/servosity/msp-skills/tree/main/skills/riverside-fm. The skill defines how its required CLI (riverside-fm-cli) can be installed via the openclaw: frontmatter block.
OpenClaw isn't generally available yet; the frontmatter wiring is pre-shipped and will activate the moment OpenClaw launches.
Authenticate
See mcp-install.md for the credentials riverside-fm-cli needs.
What this skill does
| Question you keep asking | Command |
|---|
| Back up everything in a studio to disk? | riverside-fm-cli bulk export --studio my-studio --out ./archive |
| Get the most useful asset for one recording in a single shot? | riverside-fm-cli grab <session-id> --out ./dl |
| Find a quote across my whole transcript archive? | riverside-fm-cli search "network effects" --json |
| Turn a transcript into captions for a web player? | riverside-fm-cli transcripts convert <session-id> --format srt --out ep.srt |
| Who talked the most in this episode? | riverside-fm-cli transcripts talktime <session-id> --json |
| Which takes in a studio are fully ready to edit? | riverside-fm-cli ready --studio my-studio |
| Pull every Magic Clip for a project with fresh URLs? | riverside-fm-cli clips harvest --project 69fcda9fba030a19ae93a526 --download --out ./clips |
| Refresh expiring CloudFront media links before they die? | riverside-fm-cli media refresh --project 69fcda9fba030a19ae93a526 --prefetch --out ./media |
Full command reference: guide.md. For the AI-agent operating contract (--agent, --dry-run, when to confirm before mutating), see AGENTS.md.
What makes this different
Most Riverside integrations and MCP servers proxy each question into a live API call. That's fine for one record. It dies at scale, when you're asking "find the episode where we mentioned that customer story across two years of a weekly show" or "archive every take in this studio before we cancel the plan."
This skill syncs Riverside into a local SQLite mirror with full-text search. Aggregate questions become one local query: instant, offline, and the AI sees the answer, not the raw data. Compound commands like bulk export, clips harvest, and search join across studios, projects, takes, transcripts, and clip-assets - work a stateless API wrapper can't do.
The pain this closes
Riverside records studio-quality podcast and video, but getting your own content back out is a manual slog. There is no bulk export: every transcript, audio track, and video file comes out one click at a time, per take, down the Studio then Project then Take path. Back up a fifty-episode show and you face hundreds of clicks. The friction is real enough that podcasters in community groups - such as the Podcasting subreddit and the podcast-host Facebook groups - keep asking how to "batch download Riverside recordings" to archive a series or free up storage. And the official API that would fix it is gated behind a custom-priced Business plan, so Pro, Live, and Webinar accounts get no supported way to back up, search, or re-format their own recordings.
This skill closes that gap:
bulk export - archive a whole studio (or a date range) to disk, resumable.grab - one command that downloads the most useful asset for a recording (transcript, then audio, then video).search - full-text search across every transcript in your local mirror.transcripts convert - turn a transcript into WebVTT / SRT / Markdown captions the Riverside UI never exposes.clips harvest - pull every Magic Clip for a project with freshly refreshed download URLs.
See pain-point.md for the longer narrative.
Frequently asked questions
Does this work with ChatGPT?
Yes, on Plus, Pro, Team, Business, Enterprise, and Education plans (Free tier does not yet expose Developer Mode). ChatGPT connects to remote MCP servers over HTTPS, not local stdio binaries. The Riverside MCP server is local, so for ChatGPT you expose it via the supergateway bridge or your own HTTPS endpoint. Step-by-step in mcp-install.md.
Does this work with Codex, Cursor, Windsurf, Cline, Copilot, or Gemini?
Yes - all of them speak MCP. Cross-tool install commands are in the matrix above and the deep-dive in docs/which-agent.md.
Do I need to know how to code?
No. The recommended install is to paste one sentence into Claude Code or Codex - your agent reads SKILL.md and does the install. The fallback is a one-line installer per OS (bash or PowerShell). Neither path requires writing code. You'll enter your Riverside credentials once.
Is my Riverside data safe?
Your data stays on your machine. The CLI and MCP server are local binaries. The SQLite mirror sits in a directory under your user account. The AI agent only sees what the CLI returns - typically a query result, not raw bulk data. Credentials are read from your environment or your agent's config; never bundled into this repo or transmitted anywhere by MSP Skills.
Do I need a Riverside Business plan or an API key?
No. The official Riverside API requires a custom-priced Business plan. This CLI reuses the session cookies from your already-logged-in browser to reach the same internal API the Riverside web app uses, so it works on Pro, Live, and Webinar accounts that can't issue an API key. Run riverside-fm-cli auth login --chrome once.
Can this download other people's recordings?
No. It authenticates as you, with your own browser session, and only reaches the studios and takes your account already has access to. It is scoped to exporting your own content, not scraping anyone else's.
Will this change anything in my Riverside account?
The commands you use to export, search, and convert all read - they only download and transform your data. The CLI does include the generic import command, which can POST records from a local file via the API's create/upsert path; it is not part of any workflow in this skill, and an agent should treat it as a reviewed write. Nothing in this CLI deletes or publishes.
What does it cost?
Free. Apache-2.0 licensed. You pay only for whichever AI agent you use (Claude, ChatGPT, Codex, etc.), and that's billed by your AI provider, not by us.
Safety model
The Riverside data commands in this skill read: the export, search, download, and transcript commands only GET from your account. The one mutating command is the generic import, which POSTs records from a local JSONL file - it is not part of any export workflow here.
| Tier | Examples | Recommended agent policy |
|---|
| Read | bulk export, grab, search, transcripts convert, transcripts talktime, clips harvest, media refresh, ready, wait, sync | Allow |
| Write (routine) | import (POSTs records from a local JSONL via the API's create/upsert path); the endpoint-mirror capability checks the safety scanner flagged by verb name (ai can-create-event, takes get-assets, clips get-patches) are themselves read-only GETs | Preview with --dry-run, then a reviewed write |
| Destructive / config | none - this CLI exposes no delete, publish, or account-config command | Human-in-the-loop only |
The strongest control is the scope you grant the Riverside credentials - the CLI can only do what the credentials are permitted to do. Full details, including how to lock it down, are in governance.md.
Status
Beta. Validated against the Riverside API surface and being validated with MSPs running it live against their own production tenant in our weekly Build Sessions. RSVP at compoundingteams.com/build-sessions.
Standards. Conforms to the open Agent Skills spec (Anthropic, Dec 2025; 40+ agents). MCP-compatible - works with any MCP-capable agent including Hermes. OpenClaw-ready (frontmatter pre-wired, awaiting OpenClaw launch).
Maintained by Servosity. Apache-2.0 licensed. Built with CLI Printing Press. Last updated: 2026-06-30.