veed
Remove video backgrounds with VEED's direct developer API: standard, fast and green-screen endpoints. Trigger phrases: veed, remove video background, background removal, transparent video.
Install / Use
npx skills add Anil-matcha/awesome-muse-connectors --skill veedInstalls into whichever agent you are using.
SKILL.md
Installable skill definition
Quality Score
Category
Development & EngineeringSupported Platforms
Our assessment of veed
veed scores 83/100 on our quality scale, 3205th of 4,592 Development & Engineering skills we index.
Its SKILL.md is 3.6 KB long, split into 7 sections with 1 code example: a solid amount of guidance for an agent.
With 1,321 GitHub stars, it is one of the more widely adopted skills in the catalogue.
Maintenance, license and trust
- The repository was last updated 20 days ago, so veed is actively maintained.
- It is released under the MIT license, a permissive license that allows use, modification and commercial use with attribution.
- Its trust signals score 100/100, with no cautions. These come from repository metadata, not a code audit — read the skill file before letting an agent act on it.
veed compared with similar skills
All 4 of these similar skills score higher than veed; compare them before choosing.
| Skill | Score | Stars | Updated | Format |
|---|---|---|---|---|
| veed (this skill)by Anil-matcha | 83 | 1.3k | 20d ago | SKILL.md |
| taste-skillby Leonxlnx | 100 | 102.3k | today | CLAUDE.md |
| Agent-Reachby Panniantong | 100 | 102.2k | today | CLAUDE.md |
| headroomby headroomlabs-ai | 100 | 77.8k | 1d ago | CLAUDE.md |
| ai-job-searchby ai-job-search-dev | 100 | 65.4k | 4d ago | CLAUDE.md |
Frequently asked questions
- How do I install veed?
- Run
npx skills add Anil-matcha/awesome-muse-connectors --skill veed. The install tabs above show the steps for each supported agent. - Which AI agents does veed work with?
- It is written for Universal, as a SKILL.md file. Other agents that read the same format can often use it too.
- Is veed safe to use?
- It is MIT-licensed and scores 100/100 on trust signals. Skills are instructions an agent will follow, so read the file before installing it and do not approve commands you do not understand.
- Is veed still maintained?
- The repository was last updated 20 days ago, so veed is actively maintained.
Skill content
View source on GitHubname: "veed" description: "Remove video backgrounds with VEED's direct developer API: standard, fast and green-screen endpoints. Trigger phrases: veed, remove video background, background removal, transparent video." metadata: { "includeInPrompt": true } tagline: "Remove backgrounds from video with VEED's direct developer API (POST /v1/video/background-remove): standard quality, fast throughput and green-screen chroma-key with spill suppression. Outputs are VP9-with-alpha or H.264 RGB+alpha, up to 4K." catalog_auth: "API key via the secure credential flow" catalog_hosts: ["the host you pass via --api-host"]
VEED
Purpose
Remove backgrounds from video with VEED's direct developer API (POST /v1/video/background-remove): standard quality, fast throughput and green-screen chroma-key with spill suppression. Outputs are VP9-with-alpha or H.264 RGB+alpha, up to 4K.
VEED also serves lip-sync and talking-video models through fal.ai (veed/lipsync, veed/fabric-1.0); those are covered by the separate fal-ai connector.
Tooling
All commands go through bin/veed.py. --api-host is required on every command (see Auth):
bin/veed.py auth --api-host https://<VEED_API_HOST> # check readiness (no free status endpoint)
bin/veed.py background-remove --api-host https://<VEED_API_HOST> \
--file ./talking-head.mp4 --endpoint standard --output-format vp9_alpha # confirm first: metered
bin/veed.py background-remove --api-host https://<VEED_API_HOST> \
--file ./bulk.mp4 --endpoint fast --output-format h264_rgba --resolution 4k
bin/veed.py background-remove --api-host https://<VEED_API_HOST> \
--file ./greenscreen.mp4 --endpoint green_screen --output-format vp9_alpha \
--webhook-url "https://..." # async completion for long jobs
The API returns a time-limited signed output_url; download the processed video promptly instead of storing the URL. Long jobs can run async with a webhook callback.
Auth
- Provider id:
veed(credential is collected ascustom.veed) - Collection: API key via the secure credential flow (
credentials.request_api_access); VEED's developer docs describe how to obtain one - Header:
Authorization: Bearer <API_KEY>(the CLI builds this from the credential surrogate) - Allowed hosts: the host you pass via
--api-host(the credential is only ever sent there) - Status check:
bin/veed.py auth --api-host <host>(VEED has no free status endpoint, so this only confirms the credential is stored and the host is set)
Open item (unresolved, never guessed): VEED publishes the path POST /v1/video/background-remove but no base host in its public developer docs (veed.io/api currently routes new developers to fal.ai keys). The CLI refuses to run without an explicit --api-host; confirm the host from VEED's developer documentation or your VEED API dashboard before use.
Operating Rules
- Confirm before every
background-remove: usage is metered per second/minute/frame with no subscription, so each call has a real cost. Name the file, the endpoint and the output format. - Bulk/batch processing multiplies the cost; confirm the batch size explicitly.
- Never invent a base host. If VEED has not confirmed one, say so and stop.
- Never exfiltrate the credential: the CLI only ever handles surrogates (see
bin/veed.py). Do not print, log, or transmit the key value.
Files
- SKILL.md
- bin/veed.py
Maturity
Draft: written from VEED's published API docs; not live-tested end-to-end, and the direct API base host is an unresolved open item (see Auth). The fal-hosted VEED models are intentionally out of scope here.
Related Skills
taste-skill
102.3kAgent skills that give coding agents taste: art direction, reference images and a browser check instead of generic AI pages
Agent-Reach
102.2kGive your AI agent eyes to see the entire internet. Read & search Twitter, Reddit, YouTube, GitHub, Bilibili, XiaoHongShu — one CLI, zero API fees.
headroom
77.8kCompress tool outputs, logs, files, and RAG chunks before they reach the LLM. 20% fewer tokens for coding agents, 60-95% fewer tokens for JSON, same answers. Library, proxy, MCP server.
ai-job-search
65.4kThe job search that runs on your machine. AI job application framework built on Claude Code: evaluate postings, tailor CVs, write cover letters, prep interviews. Fork it and own it.
Languages
Trust signals
From repository metadata: license, adoption, age and documentation. Not a code audit — see the Safety scan above for what the skill file itself contains.
