The full upstream README, mirrored here for reference. Install config, tool schemas, adoption signals, and an original overview live on the Zvid listing page.
Official Zvid agent and MCP server. Creator follows a quality-first workflow: plan the brief, adapt designed examples or library assets, validate exact project JSON, save a reviewable draft, then render only after approval. Automation and Developer add trusted direct operations.
There are exactly two ways to authenticate. Every setup below uses one of them:
| # | Method | Works with | What you need |
|---|---|---|---|
| 1 | OAuth — sign in with your Zvid account (recommended) | Hosted endpoint https://mcp.zvid.io/mcp | A Zvid account and any OAuth-capable Streamable HTTP MCP client. No API key to create or paste. |
| 2 | API key (zvid_…) | Hosted endpoint via the X-Api-Key header, or local stdio / self-hosted | A key from app.zvid.io/api-keys; Node.js ≥ 18 for stdio |
Method 1 — OAuth sign-in. Add https://mcp.zvid.io/mcp to your MCP client
and sign in when prompted. The endpoint publishes OAuth discovery metadata,
uses authorization code + PKCE with the single scope zvid:mcp, and issues
short-lived access tokens with rotating refresh tokens (revocable per
RFC 7009). This is what the client-specific instructions below use.
Method 2 — API key. Create a key at app.zvid.io/api-keys, then either:
X-Api-Key header on https://mcp.zvid.io/mcp —
for programmatic clients that cannot run an OAuth flow, or legacy access.ZVID_API_KEY / --api-key as
shown in Local stdio / self-hosted setup.
The stdio transport is API-key only, because MCP OAuth is defined for HTTP
transports.The hosted endpoint is https://mcp.zvid.io/mcp. It accepts both
authentication methods: OAuth sign-in (method 1 — used by the client
instructions below) and an API key in the X-Api-Key header (method 2).
You can also open /mcp inside Claude Code and authenticate there.
Add this to ~/.codex/config.toml, then authenticate from MCP settings or run
codex mcp login zvid:
The dashboard stores the default tool profile
and maximum credits per render for MCP clients. A client may connect with
concrete values in the endpoint, for example
https://mcp.zvid.io/mcp?profile=creator&maxRenderCredits=60; this is connection
configuration and cannot be changed by the model during a conversation. The
default per-render credit limit is 120, and the dashboard limit, when set, is a
hard ceiling: the effective limit is the lower of the dashboard value and the
value requested in the endpoint.
The official n8n workflow copies both dashboard defaults into its workflow JSON when downloaded. Each workflow can then choose another concrete profile and credit ceiling locally without updating the dashboard or any other workflow.
The npm/stdio entry point uses an API key (authentication method 2).
Configuration resolution is CLI flags, then environment, then
~/.zvid-mcp.json:
| Env var | CLI flag | Required | Default | Purpose |
|---|---|---|---|---|
ZVID_API_KEY | --api-key zvid_… | stdio only | — | Zvid API key |
ZVID_API_URL | --api-url http://… | no | https://api.zvid.io | Orchestrator base URL |
ZVID_MCP_PROFILE | --profile creator | no | creator | Tool profile |
ZVID_MCP_MAX_RENDER_CREDITS | — | no | 120 | Legacy/fallback hosted limit; dashboard and explicit n8n values take precedence |
ZVID_MCP_MAX_BULK_ITEMS | — | no | 25 | Automation/developer bulk-call item ceiling |
ZVID_MCP_QUOTE_SECRET | — | hosted | process-random | Shared HMAC secret for multi-instance quotes |
For a self-hosted OAuth deployment, keep the resource and issuer identical on both services:
| Service | Variable | Production default |
|---|---|---|
| Orchestrator | OAUTH_ISSUER | https://api.zvid.io |
| Orchestrator | OAUTH_MCP_RESOURCE | https://mcp.zvid.io/mcp |
| Orchestrator | OAUTH_CONSENT_URL | https://app.zvid.io/oauth/authorize |
| MCP | ZVID_MCP_RESOURCE | https://mcp.zvid.io/mcp |
| MCP | ZVID_OAUTH_ISSUER | https://api.zvid.io |
The orchestrator also accepts OAUTH_ACCESS_TOKEN_TTL_SECONDS and
OAUTH_REFRESH_TOKEN_TTL_SECONDS; their defaults are one hour and 30 days.
From a checkout:
The default creator profile is the former Advanced authoring surface. It exposes
planning, designed examples, creative-library search, stock media, schema/docs,
repair, validation, projects and templates alongside approval-aware draft tools.
Creator deliberately rejects create_media and revise_media calls that omit the
complete project payload. This prevents a weak model from improvising a poor
layout from a brief. The agent must plan, adapt or assemble an exact payload,
validate it remotely, fix every error and layout warning, and only then save the
draft. create_media remains non-spending; render_media still requires the
user-approved signed quote.
When the user asks for a reusable template rather than a one-off video or
image, create_media_template saves a parameterized project — a top-level
variables object of safe defaults referenced via {{name}} placeholders —
as a persistent tpl_… template. It returns the declared variables, a credit
estimate for rendering the untouched defaults, and an editor deep link
(https://editor.zvid.io/?template=<id>). create_media_from_template then
instantiates the template with new variable values as an approval-gated draft
with a signed quote. Neither call spends credits. In the creator profile
create_media_template requires the complete parameterized payload, mirroring
create_media.
Rendering is deliberately separate from drafting. Quote tokens are HMAC-signed, expire after 15 minutes, bind the draft ID, project version, canonical payload hash, media type and estimated credits, and are revalidated immediately before submission. An idempotency UUID prevents retry duplication.
| Profile | Intended use | Surface |
|---|---|---|
readonly | Inspection | get_media, list_media, get_account |
creator | Most users | Quality authoring, libraries, validation, projects/templates and approved renders |
automation | Trusted workflows | Creator tools plus direct renders, capped bulk rendering and webhook operations |
developer | Full API development | Every registered non-destructive/raw tool |
Update and delete tools remain disabled in every profile. Stored webhook
secrets are redacted by get_webhook; only create_webhook returns a new
secret. Bulk calls default to 25 items even though the underlying API can
accept more.
The server also publishes four user-invoked workflow prompts (product promo,
social reel, thumbnail and square post) and safe MCP resources at
zvid://authoring/guidelines and zvid://account/summary.
The following low-level capabilities are available only in the profiles described above:
| Tool | Description |
|---|---|
plan_creative_video | Plan-aware storyboard and art direction from a brief — plus libraryCandidates: published examples ranked against the brief with an adapt-vs-assemble decision. Supports consistent, fresh, and explore modes, recent-asset exclusions, brand tokens, style packs, and scene recipes |
find_matching_examples | Rank the entire published examples library against a brief (category synonyms, aspect/duration fit) and return candidates with thumbnails plus a decision: adapt-example, adapt-or-assemble, or assemble-similar (with design-template/canvas-preset/shape modules) |
start_from_example | Fetch an example's render-ready project JSON plus an adaptation map (variables, text/media slots, scene summary, fonts) and the adaptation contract — the premium path: keep the layout, swap copy/media/brand. Suggests free alternatives when a premium example is plan-locked |
render_from_example | One-call premium render: example slug + new variable values → the server saves it as a template, dry-runs the variables, and queues the render with the designed layout fully intact. The most reliable path for smaller models |
search_creative_library | Search complete examples, animated Design Studio templates, canvas presets, or shapes; results include preview/thumbnail metadata when published |
get_creative_asset | Fetch metadata and full JSON content for one creative-library item |
list_stock_providers / search_stock_media | Discover configured image/video/GIF/audio providers and search normalized render-ready media URLs |
get_project_schema | Live caller-plan-aware JSON Schema (draft 2020-12) for a project payload or full render request, plus validation notes, professional authoring guidelines, and required workflow; falls back to the bundled default schema when the endpoint is unavailable |
validate_project_json | Validate a payload before rendering — field-level errors, free. Also lints layout: overlapping texts, x/y ignored by presets, off-canvas boxes, padding cut-offs, low contrast. remote: true also runs the live API validator with your plan's real limits |
list_supported_elements | All element types (IMAGE, VIDEO, GIF, SVG, TEXT, AUDIO, SUBTITLE, SCENE) with required fields |
get_element_docs | Per-element docs: every field, constraints, gotchas, and a valid example |
get_example_payload | Validated example payloads: promo video, template render, still image, subtitles, webhook flow |
repair_project_json | Conservative auto-fix for invalid payloads with an explanation of every change |
create_render | Queue a video render from a project JSON (payload) or a template + variables |
create_image_render | Queue a still-image render (PNG/JPEG/WebP; supports snapshotTime, quality, transparent) |
get_render | Job state (waiting|active|completed|failed), progress, output url + thumbnailUrl |
list_renders | List render jobs (filter by type) |
create_bulk_render | One template/payload × N variable sets → N jobs (max 500) |
get_bulk_render / list_bulk_renders | Inspect bulk batches |
list_templates / get_template | Browse owned templates; get_template returns the full project JSON |
create_template / duplicate_template | Create a plan-validated reusable template or make an active editable copy |
preview_template | Dry-run variable resolution + validation — costs no credits |
list_projects / get_project / create_project | Editor draft projects (open at https://editor.zvid.io/?project=<id>) |
list_webhooks / create_webhook / get_webhook / test_webhook / list_webhook_deliveries | Webhook endpoints for render.completed / render.failed (HMAC-SHA256 signed) |
get_credits / get_usage_stats | Credit balance and usage |
The server doesn't just forward payloads — it knows the Zvid project schema. The schema tools are backed by a shared module (../schema, vendored as src/zvidSchema.ts) that is derived from the live backend validation (orch/middleware/validation.js) and parity-tested against it, so tool answers never drift from what the API actually accepts. When public docs and these tools disagree, the tools (backend) win.
The server also advertises this workflow as MCP instructions, so compliant clients receive it at initialize time. The example-first flow for an AI client authoring a video:
plan_creative_video with the brief, format, duration, brand and variation mode. The response includes ranked libraryCandidates and a decision.adapt-example: start_from_example with the top slug to see its variables and adaptation map, then the easiest premium path — pick new variable VALUES (copy, media URLs, brand colors) and call render_from_example { slug, variables }; the server keeps the designed layout/animations intact. Edit the payload manually only when variables cannot express the change (keep the layout skeleton; never simplify a complex example into plain text). Manual template route: create_template → preview_template → create_render { template, variables } — variables are inert on direct payload renders.assemble-similar (or nothing close on inspection): assemble the planned scene recipes from the returned design-templates, canvas-presets and shapes modules (or search_creative_library). Never force an unrelated full-video template.search_stock_media for topic-specific visuals and music for each scene — full-quality src URLs, natural size ≥ the canvas.get_project_schema / list_supported_elements / get_element_docs for the exact shape.validate_project_json (add remote: true to check against the caller's real plan) and fix every error and layout warning. Use repair_project_json only for mechanical mistakes.create_render / create_image_render, inspect a draft, revise if needed, then poll get_render for the final output.For stills or quick "make something like X" requests without a full plan, start at find_matching_examples (type: "image" for stills).
Variation behavior:
consistent derives a stable seed from the brief (or accepts variationSeed) for reproducible automations.fresh produces one new direction and expects recent library assets to be excluded.explore returns 2-5 materially different style/layout/storyboard directions, not recolors of one payload.For a reusable template, replace the final render with create_template. Later use get_template to inspect it, preview_template before rendering, and duplicate_template for a safe editable copy. Update and delete mutations are intentionally not exposed as MCP tools.
Deliveries are signed: X-Zvid-Signature: sha256=HMAC_SHA256(secret, "<X-Zvid-Timestamp>.<raw body>"). The secret (whsec_…) is returned once by create_webhook; get_webhook redacts it.
Not published yet.
npm. Bump version in package.json, then npm publish from mcp/ (public access is set via publishConfig). The prepublishOnly hook builds dist/.
Official MCP Registry (registry.modelcontextprotocol.io) — lists the server for discovery in MCP-aware clients. It reads server.json; ownership of the io.zvid/* namespace is proven with a DNS TXT record on zvid.io (or switch the mcpName/server.json name to io.github.Zvid-io/zvid-mcp to verify via GitHub instead). After npm publish:
Keep versions in sync on every release — a mismatch blocks the registry publish. Update all three together: package.json version, server.json version, and server.json packages[0].version. The server name must match in both files: package.json mcpName == server.json name.