What is media-generation?

A set of instructions for creating images, videos, and talking-head animations through AI media services such as fal.ai. It covers paid requests, long-running jobs, downloading results, and file handling.

Use it to create image-to-video clips, lip-synced talking heads, face animations, and other generated media.

It helps avoid unexpected API charges and losing track of jobs that take time to finish. It lays out a process for estimating costs, checking results, and handling large output files.

media-generation — Skill for Claude Code and Codex by S3YED

Install

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.

Any agent

npx skills add S3YED/appie-kit --skill media-generation

Clone the repo

git clone --depth 1 https://github.com/S3YED/appie-kit

Made for: Claude Code, Codex.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/skills/s3yed/appie-kit/media-generation/github.svg)](https://agentmods.dev/skills/s3yed/appie-kit/media-generation)

Your own site

<a href="https://agentmods.dev/skills/s3yed/appie-kit/media-generation"><img src="https://agentmods.dev/badge/skills/s3yed/appie-kit/media-generation/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.

Your own site · 80×15

<a href="https://agentmods.dev/skills/s3yed/appie-kit/media-generation"><img src="https://agentmods.dev/badge/skills/s3yed/appie-kit/media-generation.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session 40 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.

When invoked 1,789 The whole file, excluding the scripts and references it only reads on demand.

Security scan A 1 finding. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Token cost

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	$0.00040	$0.01789
Opus 5	$0.00020	$0.00894
Sonnet 5	$0.00008	$0.00358
Haiku 4.5	$0.00004	$0.00179

Measured 9d ago against content hash bb76a8d37533, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

media-generation scanned grade A with 1 finding 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 9d 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.

Makes network callslowCapability

Not a fault in itself. Listed so you know the mod talks to something, and to what.

curl -X POST "https://queue.fal.run/fal-ai/{app}/requests/{id}/status" \

skills/creative/media-generation/SKILL.md · 184 lines

How it starts

The opening of the file, as written. The whole thing — 184 lines — stays where its author put it; the contents beside it link to each section on GitHub.

AI Media Generation (fal.ai & Similar APIs)

When to Use

User wants to create a talking-head video from a photo + audio
User asks for image-to-video, lipsync, or face animation
Any paid API media generation task (fal.ai, etc.)
Uploading/serving generated media files

Principles

Always Estimate Costs Before Running

NEVER fire a paid API call without estimating the cost first. This is the #1 rule from Seyed after the $16 fugu-sync3 incident.

# Sync-3 v3 image-to-video: $0.1333 per output second
# 71s audio = 71 × $0.1333 = ~$9.46

Cost estimation checklist:

Look up the model's pricing (per-second, per-video, per-image)
Calculate: duration × unit_price = estimated_cost
Present to user with balance/remaining info
Only proceed after user confirms

Bulletproof Async Workflow (fal.ai)

The correct sequence for long-running fal.ai jobs is:

1. submit() → get SyncRequestHandle
2. IMMEDIATELY save handle.request_id to disk ← CARDINAL RULE
3. Poll via REST API (GET) until Completed
4. Download result video
5. Compress if needed (Telegram 50MB limit)
6. Clean up temp files

NEVER use subscribe() — it blocks until completion with no request_id fallback. If the client disconnects mid-processing, the result is unrecoverable and credits are wasted.

Request ID Recovery

Even with proper save, if the polling process dies:

The saved request_id lets you recover via REST API
Use GET on https://queue.fal.run/fal-ai/{app}/requests/{request_id}
This returns the full result JSON including video URL
The from_request_id() method in fal-client requires an httpx.Client instance (internal) — use raw REST API instead

Supported Models & Pricing

fal.ai — sync-3 Lipsync

Model	Endpoint	Pricing	Notes
sync-3 image-to-video	`fal-ai/sync-lipsync/v3/image-to-video`	$0.1333/sec output	Photo + audio → talking video
sync-3 video-to-video	`fal-ai/sync-lipsync/v3`	~$0.13/sec	Existing video + new audio

Read the full file on GitHub · 184 lines

Changes

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.

9d ago First seen · 184 lines · 40 tokens per session scan A bb76a8d37533

Subscribe to this mod's changes

media-generation is a skill published in the GitHub repository S3YED/appie-kit (9 stars, last pushed 17d ago), licensed MIT. It adds 40 tokens to every session and 1,789 once invoked, about $0.0002 per session on Opus 5. A static security scan graded it A with 1 finding (makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-03.

Creative and media agent-skills ai-agent ai-employee ai-employee-kit anthropic automation

browser-edge-cases

SOP for debugging browser automation failures on complex websites. Use when browser tools fail on specific sites like LinkedIn, Twitter/X, SPAs, or sites with Shadow DOM.

aden-hive/hive · 40 tokens

aws-patterns

Lambda best practices, S3 event patterns, SQS/SNS fanout, and DynamoDB access patterns for serverless AWS architectures.

vibeeval/vibecosystem · 31 tokens

review

Review the changes since a fixed point (commit, branch, tag, or merge-base) along two axes — Standards (does the code follow this repo's documented coding standards?) and Spec (does the code match what the originating issue/PRD asked for?). Runs both reviews in parallel sub-agents and reports them side by side. Use…

stevesolun/ctx · 95 tokens

agents-md-protocol

Create or review an AGENTS.md file so coding agents get stable repo-local instructions: environment setup, testing, style, security boundaries, PR policy, and handoff rules. Use when a repo lacks durable agent guidance or when a custom harness needs a predictable context file.

stevesolun/ctx · 59 tokens

investment-memo-generator

Investment memo creation combining financial analysis, document generation, and structured templates. Use when creating investment memos, pitch decks, deal summaries, or investment committee materials.

travisjneuman/.claude · 37 tokens

python-memory-safe-scripts

Memory-safe Python script patterns for long-running processes under systemd MemoryMax constraints. Covers allocator purge (mimalloc/glibc malloctrim), HTTP response lifecycle, DataFrame cleanup, thread-local connection reuse, and periodic GC cadence. Battle-tested through 5 OOM optimization cycles on production GPU…

terrylica/cc-skills · 197 tokens

media-generation