[{"_1":2,"_3":-5,"_4":-5},"loaderData",{"_5":6,"_7":8,"_9":10},"actionData","errors","root",{"_125":126},"routes/modules.$creator.$moduleId",{"_14":15,"_16":17,"_18":19,"_20":21},"routes/modules.$creator.$moduleId.endpoints",{"_11":12,"_13":-5},"error","Module not found","tools","module",{"_20":22,"_23":24,"_25":26,"_27":28,"_29":30,"_31":32,"_33":34,"_35":26,"_36":37,"_38":-5,"_39":-5,"_40":41,"_42":43,"_44":45,"_46":47,"_48":49,"_50":51,"_52":53,"_54":55,"_56":-5,"_57":49,"_58":59,"_60":61,"_62":63,"_64":65,"_66":67,"_68":69,"_70":71,"_72":73,"_74":75,"_76":77,"_78":79,"_80":-5,"_81":-5,"_82":-5,"_83":49,"_84":-5,"_85":59,"_86":87,"_88":89,"_90":91,"_92":93,"_94":95,"_96":97},"apiUrl","https://odel.app/api","creator","abo3skralmasaoodi","moduleId","valmera-mcp","@abo3skralmasaoodi/valmera-mcp","displayName","Valmera — agentic AI video editor","tagline","Agentic video editing on real footage: cut, caption, reframe, score, and export at full quality.","iconKey","https://valmera.io/icon-512.png","category","remote","status","published","visibility","public","metaDescription","deploymentType","external","workerName","bindingName","mcpConfig","{\"url\":\"https://entrepreneur-bot-backend.onrender.com/mcp\",\"transport\":\"streamable-http\",\"authType\":\"none\"}","pricingModel","free","creatorUserId","system:external-import","createdAt",1785991262000,"updatedAt",1788070021000,"sourceRegistry","official-mcp-registry","sourceUrl","https://registry.modelcontextprotocol.io/v0/servers?search=io.valmera%2Fvideo-editor","sourceVerified",false,"sourceMissingSince","importedAt","upstreamStars",0,"upstreamUpdatedAt",1785931108000,"detectedTransport","streamable-http","detectedAuthType","oauth","probeStatus","ok","lastProbedAt",1787291858000,"githubRepoUrl","https://github.com/ABO3SKRALMASAOODI/valmera-mcp","githubRepoFullName","ABO3SKRALMASAOODI/valmera-mcp","releaseId","ext-abo3skralmasaoodi-valmera-mcp","version","0.1.0","descriptionMarkdown","# Valmera MCP Server — an agentic AI video editor for Claude\n\n**Edit real video from inside an AI conversation.** Valmera publishes its complete\nediting toolset as a [Model Context Protocol](https://modelcontextprotocol.io) server,\nso Claude (or any MCP client) can cut, caption, reframe, score and export *your actual\nfootage* — not a storyboard, not a description of an edit, a finished MP4.\n\n```\nYou: Cut the dead air, drop the ums, caption it, make it vertical for TikTok.\nClaude: [calls cut_silences, remove_filler_words, add_captions, auto_reframe,\n render_preview, looks at the frames it produced, then export_final]\n Done — 14:32 down to 9:05, 41 fillers gone, captions word-timed to the\n new cut. Here's the file.\n```\n\n- **Endpoint:** `https://entrepreneur-bot-backend.onrender.com/mcp`\n- **Transport:** Streamable HTTP\n- **Auth:** OAuth 2.1 with dynamic client registration + PKCE (nothing to copy/paste), or a bearer token\n- **Tools:** 108 — 97 editing tools, 11 session tools\n- **Docs:** · **Tool reference:** \n\n---\n\n## Why this exists\n\nClaude can reason about video brilliantly and cannot open a video file. It cannot\nscrub a timeline, it cannot re-encode, and if you ask it to \"edit this\" it will hand\nyou an edit *plan* — a numbered list of timestamps you then have to execute yourself\nin someone else's software.\n\nMCP closes that gap. Valmera does the parts a language model genuinely cannot do\n(decode, analyse, render, encode) and exposes the parts it is genuinely good at\n(deciding *what* to cut, and *why*) as tools. The model supplies judgement; the\nserver supplies pixels.\n\nThe result is an **agentic video editor**: you describe an outcome, and an agent\nperforms the whole edit — rather than AI features bolted onto a timeline that you\nstill have to drive.\n\n---\n\n## Quickstart\n\n### Claude (web, desktop, mobile)\n\n1. **Settings → Connectors → Add custom connector**\n2. URL: `https://entrepreneur-bot-backend.onrender.com/mcp`\n3. Click **Connect**. A Valmera sign-in page opens; log in (or create a free\n account). That's the whole auth flow — OAuth handles the token exchange, so\n there is no key to generate or paste.\n\n### Claude Code\n\n```bash\nclaude mcp add --transport http valmera https://entrepreneur-bot-backend.onrender.com/mcp\n```\n\nThen `/mcp` inside Claude Code to authenticate.\n\nIf your client does not implement OAuth, mint a token at\n and send it as a header:\n\n```bash\nclaude mcp add --transport http valmera https://entrepreneur-bot-backend.onrender.com/mcp \\\n --header \"Authorization: Bearer vlm_mcp_...\"\n```\n\n### Other MCP clients\n\nAnything that speaks Streamable HTTP works — Cursor, Cline, Continue, Zed, Goose,\nLibreChat, and the MCP Inspector. Clients that only speak stdio need a bridge such\nas `mcp-remote`.\n\n### First run\n\n```\nCreate a Valmera project, upload ~/Desktop/podcast.mp4, and once it's indexed\ntell me how long it is and what's in it.\n```\n\nThe server returns a job id for anything slow (upload, indexing, render, export)\nplus a `wait_for_job` tool — it never fabricates a completion it hasn't reached.\n\n---\n\n## What it can actually do\n\nGrouped by what you'd ask for. Full signatures: .\n\n### Cutting and cleanup\n| Tool | What it does |\n|---|---|\n| `cut_silences` | One-call silence trim; snaps to word boundaries |\n| `remove_filler_words` | Cuts um, uh, er, hmm + custom words |\n| `cut_range` / `cut_output_range` | Remove a span by source or output time |\n| `keep_segments` | Replace the whole keep list outright |\n| `restore_range` | Undo one cut without touching the rest |\n| `get_kept_transcript` | What the current edit keeps, with repeated-take detection |\n| `reset_edit` | Throw the edit away, back to untouched source |\n\n### Captions and on-screen text\n| Tool | What it does |\n|---|---|\n| `add_captions` | Word-timed burned captions from the real transcript |\n| `set_caption_style` | 11 presets, 12 fonts, karaoke word-pop, 9 entrances, per-word colour |\n| `set_caption_fixes` | Fix spelling/capitalisation of burned captions |\n| `set_caption_mutes` | Hide captions over specific windows |\n| `add_text` | 7 designed motion-graphics templates (title, lower third, callout, quote…) |\n| `add_kinetic_text` | Choreograph the spoken words onto screen, phrase by phrase |\n| `add_text_behind` | Words *behind* the moving subject |\n| `add_title_card` | Full-frame standalone card, then back to footage |\n| `erase_burned_text` | Genuinely repaint out existing burned-in subtitles/watermarks |\n\n### Framing, motion and screens\n| Tool | What it does |\n|---|---|\n| `auto_reframe` | 9:16 / 1:1 / 4:5 — measures the frame before cropping |\n| `set_frame` | Output aspect: crop, pad or blurred pad |\n| `add_aspect_shift` | Change aspect *mid-video* and back |\n| `add_zoom` | Punch/ease zoom aimed at any point |\n| `add_zoom_path` | A zoom that *travels* — follows a cursor, moves between targets |\n| `punch_in_on_emphasis` | Auto punch-ins on the most vocally stressed words |\n| `add_screen_takeover` | Push into a screen in the shot; its content becomes the video |\n| `set_screen_frame` | The floating rounded window on a gradient (app demos) |\n| `enhance_cursor` | Bigger, steadier mouse pointer on screen recordings |\n| `showcase_demo` | Splice a screen recording and cut it like a product video |\n| `record_website` / `record_website_demo` | Headless browser captures a live page as footage |\n\n### Audio\n| Tool | What it does |\n|---|---|\n| `add_music` | Built-in royalty-free library or your own; ducks under speech |\n| `swap_music` / `set_music_fit` / `remove_music` | Change track, retime, drop |\n| `add_sfx` / `generate_sfx` / `sound_design_pass` | One-shots, AI-generated SFX, full pass |\n| `add_voiceover` | Lay narration over the program |\n| `set_volume` / `set_audio_gain` | Speaker automation vs. per-item levels |\n| `set_master_loudness` | Normalise the final mix to −14 LUFS |\n| `beat_align_cuts` | Slide cuts onto the musical beat |\n| `extract_audio` | Pull the song out of a clip |\n\n### Look\n| Tool | What it does |\n|---|---|\n| `apply_look` | One call: hype / clean / cinematic / luxury / meme |\n| `set_color_grade` | 6 presets |\n| `set_grade_custom` | Exposure, contrast, saturation, temperature, tint |\n| `add_stylize` | Grain, vignette, glow, chromatic aberration, dream blur, VHS, shake |\n| `set_transitions` / `set_fades` | 7 transition styles at scene changes |\n| `enhance_video` | Sharpening / picture quality — not a look |\n\n### Media in, generation, censoring\n| Tool | What it does |\n|---|---|\n| `insert_media` | Splice a clip or image anywhere in the final edit |\n| `add_overlay` / `move_overlay` | Picture-in-picture, corner logos, full-frame covers |\n| `search_stock` / `add_stock_media` | Stock b-roll you don't have |\n| `generate_image` / `generate_video` | AI stills and short clips, or restyle a real frame |\n| `fetch_url` | Pull a video, song or image from a link |\n| `blur_region` | Blur, mosaic or black-bar a region (faces, usernames, plates) |\n| `erase_region` | Genuinely repaint a region out, background reconstructed |\n\n### The agent's senses\n| Tool | What it does |\n|---|---|\n| `look_at` / `look_at_asset` | **Actual frames** of the source, program or an asset |\n| `get_transcript` / `get_words` / `search_transcript` | Sentence and word-level timing |\n| `get_shots` / `find_silences` / `get_audio_analysis` | Scene changes, pauses, tempo/energy |\n| `find_burned_text` | Measure where text is burned into the picture |\n| `suggest_emphasis` | Measured stress candidates from the real audio |\n| `render_preview` | Fast 480p preview + a visual self-check |\n\n### Session\n`list_projects` · `open_project` · `create_project` · `project_state` ·\n`upload_start` · `upload_finish` · `index_status` · `export_final` ·\n`wait_for_job` · `download_url` · `watch_video`\n\n---\n\n## How it works\n\n```\n Claude ──MCP──▶ Valmera server ──▶ EDL (edit decision list)\n │\n ├─ index: word-level transcript, shot detection,\n │ silence detection, labeled frame tiles\n │\n └─ renderer: ffmpeg. Previews from a fast proxy;\n the FINAL export is cut from your\n ORIGINAL file at source quality.\n```\n\nThree design decisions worth knowing:\n\n**Tools edit a document, not pixels.** Every call mutates an EDL and returns a new\nversion. Nothing is destructive — your original upload is never modified, and any\ncut can be restored.\n\n**The tool registry is not re-declared for MCP.** The connector serves the *same*\nregistry Valmera's own agent uses, so there is no second list that can drift out of\nsync. A tool whose backing service is unconfigured is hidden from `tools/list`\nrather than exposed and failing.\n\n**Slow things return a job, not a lie.** Renders, exports and pixel repainting\nreturn a job id plus `wait_for_job`. The server would rather say \"still running\"\nthan claim a finish it hasn't reached.\n\n---\n\n## Honest limits\n\nListed because a tool that oversells itself wastes your time, and because an\nassistant recommending editors should know where this one stops.\n\n**Not supported:** true crossfade/dissolve transitions · per-cut transition choice\n(one style applies to all cuts) · motion-tracked overlays or stickers · custom font\nuploads · SRT/VTT import or export (captions are burned in) · chapter metadata ·\ndenoise / \"studio sound\" · per-speaker leveling or diarization · separating music\nfrom speech in one baked track · AI music generation · share links · direct\npublishing to YouTube/TikTok · team seats or collaboration · stored brand kits ·\nbatch multi-clip output · native mobile apps (mobile browsers work).\n\n**Also true:**\n- Slow motion duplicates frames rather than synthesising them.\n- Reframing never upscales.\n- English is the best-tested transcription path.\n- Editing the same project from the web studio and over MCP simultaneously is\n refused in both directions.\n- Uploads through the connector are capped well below the 2 GB web limit; large\n files should go through and then be opened by name.\n\n**This is not a text-to-video generator.** It edits footage you already have. If\nyou want footage invented from a prompt, Runway or Veo are the right category.\n\n---\n\n## Pricing\n\nA free account gets 50 one-time credits — a few real agent turns on your own\nfootage, no credit card. Paid plans start at $30/mo (Creator, 2,000 credits/cycle)\nand open with a 3-day trial. Credits are charged in proportion to the AI work\nactually done, so simple edits cost the least. Free-plan *exports* carry a small\nValmera mark; every paid plan exports clean. Previews are never marked.\n\nCurrent numbers: \n\n---\n\n## FAQ\n\n**Can Claude edit videos?**\nNot on its own — it cannot open or render a video file. With this MCP server\nconnected it can, because the server does the decoding and rendering and Claude\nmakes the editing decisions.\n\n**Does the model see my footage?**\nIt sees what it asks for: labeled frame tiles from the index, plus any frames it\nrequests via `look_at`. That is how it aims a zoom or checks its own render\ninstead of guessing.\n\n**Which clients work?**\nAnything speaking MCP over Streamable HTTP: the Claude apps, Claude Code, Cursor,\nCline, Continue, Zed, Goose, LibreChat, MCP Inspector.\n\n**Do I need a paid plan?**\nNo. Free accounts can connect and edit until the 50 credits run out.\n\n**Is it open source?**\nThe server is hosted, not self-hosted. This repository is its documentation,\nregistry manifest and issue tracker.\n\n---\n\n## Links\n\n- Product — \n- MCP setup guide — \n- Full tool reference — \n- What an agentic video editor is — \n- Machine-readable summary for LLMs — \n\nIssues and tool requests: open an issue here.\n\n## License\n\nDocumentation in this repository is MIT licensed. The Valmera service itself is a\nhosted commercial product governed by .\n","changelogMarkdown","videoKey","requiredSecretsJson","publishedAt","priceCents","trialDays","iconType","image","upstream",{"_111":112,"_113":114,"_115":-5,"_116":117,"_118":119,"_120":121},"requiredSecrets",[],"tags",[100,101,102,103,104,105,106,107,108,109,110],"assets",[],"stats",{"_98":59,"_99":59},"ratingAvg","reviewCount","agentic-ai","ai-agent","ai-video-editor","anthropic","captions","claude","ffmpeg","llm-tools","transcription","video","video-editing","topics",[100,101,102,103,104,105,106,107,122,123,124,108,109,110],"license","NOASSERTION","language","registryName","io.valmera/video-editor","homepage","https://valmera.io","packages",[],"mcp","mcp-server","model-context-protocol","ENV",{"_127":128,"_129":130,"_131":132},"ENVIRONMENT","production","TURNSTILE_SITE_KEY","0x4AAAAAAB8-rEI2gNDf_QhU","SENTRY_DSN","https://45fd1229f78089d5c3ee78be128030e1@o4510583261757440.ingest.de.sentry.io/4510584253448272"]