Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx skills add elvniv/klo-mcp --skill edit-videogit clone --depth 1 https://github.com/elvniv/klo-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/skills/elvniv/klo-mcp/edit-video)<a href="https://agentmods.dev/skills/elvniv/klo-mcp/edit-video"><img src="https://agentmods.dev/badge/skills/elvniv/klo-mcp/edit-video/github.svg" alt="Measured on agentmods" height="20"></a>Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.
<a href="https://agentmods.dev/skills/elvniv/klo-mcp/edit-video"><img src="https://agentmods.dev/badge/skills/elvniv/klo-mcp/edit-video.svg" alt="Reviewed on agentmods" width="80" height="20"></a>What it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5.1 | $0.00057 | $0.00681 |
| Opus 5.5 | $0.00023 | $0.00272 |
| Sonnet 5.5 | $0.00011 | $0.00136 |
| Haiku 4.5 | $0.00006 | $0.00068 |
Grade A, and why
edit-video scanned grade A with 0 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 4d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Nothing flagged
None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.
How it starts
The opening of the file, as written. The whole thing — 32 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Use the connected KLO Editing tools for the requested video project. Preserve the user's chosen duration, aspect ratio, style, audio, and approval decisions. Explain results in ordinary language; technical IDs belong in tool calls unless needed to resolve ambiguity.
Connect and bring footage
If KLO is not connected, direct the user to connect Klo Editing and complete its normal browser OAuth sign-in. The endpoint is https://mcp.getklo.com/mcp/editing. Do not ask for an API key or a password in chat.
Create a project when a new edit is requested, or inspect the specified existing project. Keep its projectId explicit in later calls: another conversation can change the account's current project.
When the assistant cannot access an attachment's original bytes, call create_upload_page and give its private link to the user. It opens a file picker, so the user need not know the filename or byte count. When they return, poll get_media in the same project. The upload imports automatically; do not import it twice. A directly downloadable authorized URL can instead be imported with add_footage. A local filesystem path on the user's computer is not a path on KLO's server.
Edit and revise
Wait until source media is ready. Use actual duration and dimensions from the tools. Place the requested source interval with add_media_to_timeline; preserve the source voice with linkAudio:true unless the user wants it removed. KLO times are seconds.
Use the exposed tools for individual cuts, titles, picture and sound adjustments. For static text, set delivery:"none" and animation:"none"; inspect the resulting size and position. This connector can use supplied text but does not expose automatic transcription or automatic captions. Do not invent spoken words.
Inspect project state when revising, and apply changes to that project instead of rebuilding it. undo and redo reverse saved edits. Verify the resulting state before claiming that a change was reversed or restored. Avoid additional styling, music, or transitions unless they serve the user's brief.
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 4d ago First seen · 32 lines · 57 tokens per session scan A 3bfe2c59406d
edit-video is a skill published in the GitHub repository elvniv/klo-mcp (0 stars, last pushed 5d ago), licensed MIT. It adds 57 tokens to every session and 681 once invoked, about $0.0002 per session on Opus 5.5. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-11-04.
edit-video compared with skills from other repositories
| Mod | Type | Stars | Last push | Licence | Scan | Tokens | Works with |
|---|---|---|---|---|---|---|---|
| edit-videoelvniv/klo-mcp | Skill | 0 | 5d ago | MIT | A | 57 | Claude Code |
| apple-photossweetrb/apple-photos-mcp | Skill | 45 | 4d ago | MIT | A | 158 | Claude Code |
| remarcmetedata/Remarc | Skill | 83 | 1mo ago | MIT | A | 41 | Claude Code, Codex |
| codex-bridgebuidangminh23/codex-mcp-bridge | Skill | 32 | today | MIT | A | 29 | Claude Code, Codex |
| actvtoye-collective/actvt-claude-plugin | Skill | 0 | 9d ago | MIT | A | 95 | Claude Code |
| mailbridgepooyanmajd/pmtechdev-mcp-plugins | Skill | 0 | 4d ago | MIT | A | 64 | Claude Code, Codex |
| make-demo-videoScreenifyStudio/screenify-mcp | Skill | 0 | 11d ago | MIT | A | 90 | Claude Code |