Skip to main content
Glama
hello-emrah

Svetovid

by hello-emrah
OverviewSchemaSourceRelated ServersDiscussionsAdmin
Python
Local

Svetovid

An MCP server for seeing and editing video in DaVinci Resolve Studio.

Svetovid is the four faced Slavic god who watches every direction at once, the all seer. Svet is light, vid is sight, the same ancient root that gives us the word video. The name signals the one thing most tooling cannot do: see.

Svetovid is a Model Context Protocol server for DaVinci Resolve Studio. Most tools drive an editor blind, pushing metadata they cannot see. Svetovid renders the frame at the playhead and hands it back, so the agent works from what is really on screen, not a guess. It speaks to Resolve over its scripting API, and you and the agent share the same open project. Built for personal use, shared openly, not productised.

Runs on your machine, alongside your open Resolve project.

Tools

Eighty nine tools spanning the Resolve scripting API, from orienting in a project to delivering the final file. The two at the top are the heart: Svetovid renders the real frame so the agent edits against what is on screen, not blind metadata. Everything destructive sits behind a confirm gate, the tool previews the action and only applies it when called again with confirm set, so a first exploratory call never deletes work or starts a render.

Group

Tools

See and navigate

get_state, render_current_frame, open_page, get_playhead, set_playhead_timecode, set_playhead_frame, get_current_item

Project

list_projects, get_project_info, create_project, load_project, save_project, get_project_settings, set_project_setting

Media

browse_storage, import_media, list_media_pool, create_bin, set_current_bin, get_clip_properties, set_clip_property, set_clip_mark_in_out, delete_media_clips

Timeline

list_timelines, create_empty_timeline, set_current_timeline, rename_timeline, duplicate_timeline, delete_timeline, get_timeline_settings, set_timeline_setting, set_timeline_start_timecode, add_track, delete_track, set_track_name, set_track_enabled, set_track_locked

Edit and assembly

create_timeline_from_clips, append_to_timeline, insert_title, insert_fusion_title, insert_generator, create_compound_clip, create_fusion_clip, detect_scene_cuts, delete_timeline_items

Clip transforms

list_timeline_items, get_item_properties, set_item_property, set_item_transform, set_item_enabled, set_item_clip_color, smart_reframe, stabilize, add_fusion_comp

Markers

add_marker, list_markers, delete_marker, delete_markers_by_color

Colour

list_color_groups, add_color_group, assign_to_color_group, get_node_graph_info, set_node_lut, set_node_enabled, set_cdl, copy_grade, reset_grades, grab_still

Audio

insert_audio_at_playhead, transcribe_audio, clear_transcription, set_voice_isolation, create_subtitles_from_audio, auto_sync_audio

Render and deliver

list_render_presets, load_render_preset, get_render_formats, get_render_codecs, set_render_format_codec, set_render_settings, add_render_job, list_render_jobs, start_render, stop_render, get_render_status, delete_render_jobs, list_quick_export_presets, quick_export

Every tool returns plain data the agent can read, and reaches Resolve through the shared scaffolding in connection.py (the live handle), gate.py (the confirm gate) and serialize.py (objects to plain dicts). One module per group under src/svetovid/tools/.

Related MCP server: video-editor

Requirements

  • DaVinci Resolve Studio 18.5 or newer, running. The scripting API is Studio only.

  • External scripting enabled: Resolve, Preferences, System, General, External scripting using set to Local.

  • Python 3.10 or newer.

Install

git clone https://github.com/hello-emrah/svetovid-mcp.git
cd svetovid-mcp
pip install -e .

Wire into Claude Code

Add an entry to ~/.claude.json under mcpServers:

{
  "mcpServers": {
    "svetovid": {
      "command": "svetovid-mcp"
    }
  }
}

Restart Claude Code. The tools appear under the mcp__svetovid__* namespace.

Environment

Resolve's scripting paths are auto-detected per platform. Set RESOLVE_SCRIPT_API and RESOLVE_SCRIPT_LIB in a .env only for a non-standard install (see .env.example).

Why "Svetovid"

Svetovid stood on the island of Rügen as a four faced idol, looking north, south, east and west so nothing could pass him unseen, the god of light, sight and divination. The name carries the whole idea of this tool: an editor the agent can see through, watching the timeline from every side. And the root underneath it, vid, to see, is the same one that surfaces in the word video.

Design philosophy

The visual mark and the tool itself were built deliberately against the visual language of capitalist software design. No gradients, no neon, no glass, no drop shadows, no isometric stock illustration. Single-shade flat seals in warm earth tones, ancient-glyph silhouettes, generous whitespace. The mark could be pressed into wax or carved into stone.

This tool is built for personal use and shared openly. It is not productised, monetised, or instrumented. Use it for your own work or fork it for yours.

License

MIT, see LICENSE.

Available Tools

2 tools
get_stateA

Orient in DaVinci Resolve. Call this first.

Returns the product and version, the open project, every timeline in the project, and for the current timeline its frame rate, resolution, track counts, clip count and frame range.

ParametersJSON Schema
NameRequiredDescriptionDefault

No parameters

TDQS

A4.4/5.0
Behavior4/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations exist, so the description bears full burden. It discloses the tool reads state and returns detailed information (product, version, project, timelines, current timeline metrics). Missing minor details like prerequisites (e.g., open project) or error conditions, but overall good transparency.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

The description is extremely concise with two sentences that front-load the key instruction 'Call this first' and list return data efficiently. Every sentence earns its place.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given the tool's simplicity (no parameters, no output schema), the description adequately covers its core function of returning state. It could mention potential failure cases (e.g., no open project), but overall is sufficiently complete for an agent.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters4/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

No parameters are defined, so the description adds no parameter detail. Baseline score of 4 is appropriate since schema coverage is 100% and no parameters exist.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool's purpose: to orient in DaVinci Resolve and return state information. It uses specific verbiage like 'Call this first' and lists actionable return data, distinguishing it from the sibling tool 'render_current_frame'.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines4/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly advises 'Call this first,' providing clear guidance on when to use the tool. While it doesn't explicitly state when not to use it or discuss alternatives, the context of a single sibling tool makes the usage clear.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

render_current_frameA

See the edit. Export the frame at the playhead and return it as an image.

Use this to look at what the timeline actually shows before and after an edit, rather than reasoning from metadata alone. The current timeline must be open and a frame visible at the playhead.

ParametersJSON Schema
NameRequiredDescriptionDefault

No parameters

TDQS

A4.2/5.0
Behavior3/5

Does the description disclose side effects, auth requirements, rate limits, or destructive behavior?

No annotations are provided, so the description carries full behavioral burden. It discloses that the tool exports an image and requires specific conditions, but does not mention potential delays, error handling (e.g., if no frame), output format details, or side effects. Adequate but could be more transparent.

Agents need to know what a tool does to the world before calling it. Descriptions should go beyond structured annotations to explain consequences.

Conciseness5/5

Is the description appropriately sized, front-loaded, and free of redundancy?

Three concise sentences with no wasted words. The main action ('Export the frame...') is front-loaded, and the rest provides context. Every sentence serves a purpose.

Shorter descriptions cost fewer tokens and are easier for agents to parse. Every sentence should earn its place.

Completeness4/5

Given the tool's complexity, does the description cover enough for an agent to succeed on first attempt?

Given no output schema, the description covers the essentials: what it does, when to use, and prerequisites. It lacks explicit mention of return format or error states, but for a simple tool with no parameters, it is largely complete.

Complex tools with many parameters or behaviors need more documentation. Simple tools need less. This dimension scales expectations accordingly.

Parameters3/5

Does the description clarify parameter syntax, constraints, interactions, or defaults beyond what the schema provides?

The input schema has no parameters and schema description coverage is 100% (trivially). Per the rule, baseline is 3. The description adds no parameter information because there are no parameters, so score 3.

Input schemas describe structure but not intent. Descriptions should explain non-obvious parameter relationships and valid value ranges.

Purpose5/5

Does the description clearly state what the tool does and how it differs from similar tools?

The description clearly states the tool exports the current frame as an image, using a specific verb ('Export') and resource ('frame at the playhead'). It also distinguishes itself from the sibling 'get_state' by emphasizing visual evidence over metadata.

Agents choose between tools based on descriptions. A clear purpose with a specific verb and resource helps agents select the right tool.

Usage Guidelines5/5

Does the description explain when to use this tool, when not to, or what alternatives exist?

Explicitly states when to use ('to look at what the timeline actually shows before and after an edit, rather than reasoning from metadata alone') and provides a prerequisite ('timeline must be open and a frame visible'). This effectively guides the agent on when to invoke this tool vs. alternatives.

Agents often have multiple tools that could apply. Explicit usage guidance like "use X instead of Y when Z" prevents misuse.

Tool Schema Changelog

Recent tool additions, removals, and schema changes observed during successful MCP inspections.

  1. 2 tool updatesv0.1.0
    • First observedget_state
    • First observedrender_current_frame
Deploy Server
One endpoint for every MCP toolRayrun runs every MCP server behind one URL and holds the credentials. Each call is decided per client, tool, and argument, then recorded.ray.runAd

TDQS

A4.1/5.0

Scored across 2 tools

Disambiguation5/5

The two tools serve clearly distinct purposes: one provides project and timeline metadata, the other exports a frame as an image. There is no overlap or ambiguity.

Naming Consistency5/5

Both tools follow a consistent verb_noun snake_case pattern: get_state and render_current_frame. Naming is uniform and predictable.

Tool Count2/5

With only 2 tools for a complex domain like DaVinci Resolve, the server feels significantly underdeveloped. While it covers basic inspection and frame export, many essential operations (e.g., editing, timeline manipulation) are missing.

Completeness2/5

The server provides only state retrieval and frame rendering, lacking any editing or modification capabilities. For a tool claiming to 'orient' and 'see the edit,' the absence of editing tools makes it incomplete for interactive workflows.

Maintenance

ActivityStale
ResponsivenessNo issues

Resources

Related MCP Connectors

Related MCP Servers