Text To Video Kapwing

Get text-based videos ready to post, without touching a single slider. Upload your text prompts (TXT, DOCX, MP4, MOV, up to 500MB), say something like "turn...

Security Scan

The skill broadly matches a text→video service, but there are naming/metadata inconsistencies and small scope surprises (external token minting, implied config-path access) that deserve confirmation before use.

Purpose & Capability

Name and description advertise 'Kapwing' style text-to-video, which aligns with the runtime instructions calling a remote render API. However the actual backend endpoints are on mega-api-prod.nemovideo.ai (not kapwing.com), and the SKILL.md frontmatter lists a config path (~/.config/nemovideo/) while the registry metadata shows none — this mismatch suggests either mislabeling or sloppy metadata.

Instruction Scope

Instructions stay within the stated purpose (upload text/media, create session, SSE editing, export/polling). They also instruct the agent to: mint an anonymous token from an external endpoint if NEMO_TOKEN is absent, read the skill file's YAML frontmatter for attribution, and detect install path to set X-Skill-Platform (may require inspecting typical install locations under the user's home). Those actions are explainable but widen the scope to include contacting an external auth endpoint and reading local install/config paths.

Install Mechanism

Instruction-only skill with no install spec or code files — nothing is written to disk by an installer. Low install risk.

Credentials

Only one required env var (NEMO_TOKEN) is declared, which is appropriate for an API-backed video service. Concerns: (1) SKILL.md will request an anonymous token from the network if no NEMO_TOKEN is present (network call that mints a token tied to a generated UUID), and (2) the frontmatter references a config path (~/.config/nemovideo/) that could contain credentials — this is inconsistent with the registry metadata that lists no config paths. Confirm whether the skill legitimately needs to read that path and what the token scopes are before supplying real/privileged credentials.

Persistence & Privilege

Skill does not request always: true and has no install hooks. Autonomous invocation is allowed (platform default) but not combined with other high privileges. The skill does instruct server-side job creation and polling, which is expected for this function.

What to consider before installing

This skill appears to implement a text→video service, but there are a few things to check before installing or providing credentials: - Verify the backend domain: the SKILL.md calls mega-api-prod.nemovideo.ai, yet the skill name references 'Kapwing' — confirm the provider and trustworthiness of nemovideo.ai (is this an official integration or a third-party proxy?). - Understand NEMO_TOKEN scope: only one env var is requested. Ask the publisher what permissions that token grants and whether you can use a limited-scope/test token. Avoid supplying broad or long-lived credentials until you confirm scope. - Anonymous token behavior: if you don't provide NEMO_TOKEN, the skill will contact an external endpoint to mint a 7‑day token using a generated UUID. If you prefer not to have the agent call that endpoint automatically, require the token externally or disable the skill. - Config-path inconsistency: SKILL.md frontmatter references ~/.config/nemovideo/ but registry metadata lists none. Ask whether the skill will read files from that path and why. - Attribution/file access: the skill will read its own frontmatter and inspect install paths to populate X-Skill-Platform headers — this requires the agent to look at common install locations under your home directory. If that is a privacy concern, request a version that does not attempt platform detection or denies local path inspection. If you cannot verify the above, run the skill in a restricted/test environment (or with a throwaway limited token) rather than giving it production credentials.

Getting Started

Send me your text prompts and I'll handle the AI video creation. Or just describe what you're after.

Try saying:

• "convert a 150-word product description or blog paragraph into a 1080p MP4"
• "turn this text into a short video with visuals and captions"
• "converting written content into shareable social media videos for content creators and marketers"

Getting Connected

Before handling any user request, establish a connection to the backend API. Show a brief status like "Connecting...".

If NEMO_TOKEN is in the environment, use it directly and create a session. Otherwise, acquire a free starter token:

• Generate a UUID as client identifier
• POST to https://mega-api-prod.nemovideo.ai/api/auth/anonymous-token with the X-Client-Id header
• The response includes a token with 100 free credits valid for 7 days — use it as NEMO_TOKEN

Then create a session by POSTing to https://mega-api-prod.nemovideo.ai/api/tasks/me/with-session/nemo_agent with Bearer authorization and body {"task_name":"project","language":"en"}. The session_id in the response is needed for all following requests.

Tell the user you're ready. Keep the technical details out of the chat.

Text to Video Kapwing — Convert Text Into Shareable Videos

Send me your text prompts and describe the result you want. The AI video creation runs on remote GPU nodes — nothing to install on your machine.

A quick example: upload a 150-word product description or blog paragraph, type "turn this text into a short video with visuals and captions", and you'll get a 1080p MP4 back in roughly 1-2 minutes. All rendering happens server-side.

Worth noting: shorter text blocks under 100 words produce cleaner, more focused videos.

Matching Input to Actions

User prompts referencing text to video kapwing, aspect ratio, text overlays, or audio tracks get routed to the corresponding action via keyword and intent classification.

User says...	Action	Skip SSE?
"export" / "导出" / "download" / "send me the video"	→ §3.5 Export	✅
"credits" / "积分" / "balance" / "余额"	→ §3.3 Credits	✅
"status" / "状态" / "show tracks"	→ §3.4 State	✅
"upload" / "上传" / user sends file	→ §3.2 Upload	✅
Everything else (generate, edit, add BGM…)	→ §3.1 SSE	❌

Cloud Render Pipeline Details

Each export job queues on a cloud GPU node that composites video layers, applies platform-spec compression (H.264, up to 1080x1920), and returns a download URL within 30-90 seconds. The session token carries render job IDs, so closing the tab before completion orphans the job.

All calls go to https://mega-api-prod.nemovideo.ai. The main endpoints:

1. Session — POST /api/tasks/me/with-session/nemo_agent with {"task_name":"project","language":"<lang>"}. Gives you a session_id.
2. Chat (SSE) — POST /run_sse with session_id and your message in new_message.parts[0].text. Set Accept: text/event-stream. Up to 15 min.
3. Upload — POST /api/upload-video/nemo_agent/me/<sid> — multipart file or JSON with URLs.
4. Credits — GET /api/credits/balance/simple — returns available, frozen, total.
5. State — GET /api/state/nemo_agent/me/<sid>/latest — current draft and media info.
6. Export — POST /api/render/proxy/lambda with render ID and draft JSON. Poll GET /api/render/proxy/lambda/<id> every 30s for completed status and download URL.

Formats: mp4, mov, avi, webm, mkv, jpg, png, gif, webp, mp3, wav, m4a, aac.

Skill attribution — read from this file's YAML frontmatter at runtime:

• X-Skill-Source: text-to-video-kapwing
• X-Skill-Version: from frontmatter version
• X-Skill-Platform: detect from install path (~/.clawhub/ → clawhub, ~/.cursor/skills/ → cursor, else unknown)

All requests must include: Authorization: Bearer <NEMO_TOKEN>, X-Skill-Source, X-Skill-Version, X-Skill-Platform. Missing attribution headers will cause export to fail with 402.

Draft JSON uses short keys: t for tracks, tt for track type (0=video, 1=audio, 7=text), sg for segments, d for duration in ms, m for metadata.

Example timeline summary:

Timeline (3 tracks): 1. Video: city timelapse (0-10s) 2. BGM: Lo-fi (0-10s, 35%) 3. Title: "Urban Dreams" (0-3s)

Backend Response Translation

The backend assumes a GUI exists. Translate these into API actions:

Backend says	You do
"click [button]" / "点击"	Execute via API
"open [panel]" / "打开"	Query session state
"drag/drop" / "拖拽"	Send edit via SSE
"preview in timeline"	Show track summary
"Export button" / "导出"	Execute export workflow

SSE Event Handling

Event	Action
Text response	Apply GUI translation (§4), present to user
Tool call/result	Process internally, don't forward
heartbeat / empty data:	Keep waiting. Every 2 min: "⏳ Still working..."
Stream closes	Process final response

~30% of editing operations return no text in the SSE stream. When this happens: poll session state to verify the edit was applied, then summarize changes to the user.

Error Handling

Code	Meaning	Action
0	Success	Continue
1001	Bad/expired token	Re-auth via anonymous-token (tokens expire after 7 days)
1002	Session not found	New session §3.0
2001	No credits	Anonymous: show registration URL with ?bind=<id> (get <id> from create-session or state response when needed). Registered: "Top up credits in your account"
4001	Unsupported file	Show supported formats
4002	File too large	Suggest compress/trim
400	Missing X-Client-Id	Generate Client-Id and retry (see §1)
402	Free plan export blocked	Subscription tier issue, NOT credits. "Register or upgrade your plan to unlock export."
429	Rate limit (1 token/client/7 days)	Retry in 30s once

Tips and Tricks

The backend processes faster when you're specific. Instead of "make it look better", try "turn this text into a short video with visuals and captions" — concrete instructions get better results.

Max file size is 500MB. Stick to TXT, DOCX, MP4, MOV for the smoothest experience.

Export as MP4 for widest compatibility across social platforms.

Common Workflows

Quick edit: Upload → "turn this text into a short video with visuals and captions" → Download MP4. Takes 1-2 minutes for a 30-second clip.

Batch style: Upload multiple files in one session. Process them one by one with different instructions. Each gets its own render.

Iterative: Start with a rough cut, preview the result, then refine. The session keeps your timeline state so you can keep tweaking.