production-team

production-team is an agent for Claude Code from MelonS/MelonS-Agents. It costs 76 tokens per session (895 once invoked), scanned A, original, MIT.

A production stage that turns research data into a rendered 60-second vertical video and can render it again after legal review. It supports different profile styles and an optional subject overlay for idol-focused content.

In plain words
What is it for?
Use it to render a short video, script, sources, disclosures, and a caption-verification image from research data, then apply required fixes and render again in the same mission directory.
Why use it?
It provides a repeatable way to turn researched material into a finished short while preserving source lists and disclosure files required by the legal checks.

Agent for Claude Code

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.

agentmods
npx agentmods add agents/melons/melons-agents/production-team
Clone the repo
git clone --depth 1 https://github.com/MelonS/MelonS-Agents

Made for: Claude Code.

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.

agentmods badge for production-team

README.md
[![agentmods](https://agentmods.dev/badge/agents/melons/melons-agents/production-team.svg)](https://agentmods.dev/agents/melons/melons-agents/production-team)
Your own site
<a href="https://agentmods.dev/agents/melons/melons-agents/production-team"><img src="https://agentmods.dev/badge/agents/melons/melons-agents/production-team.svg" alt="Measured on agentmods" height="20"></a>
Per session 76 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 895 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 0 findings. Scan, not verified.
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.

ModelPer sessionOnce invoked
Fable 5 $0.00076 $0.00895
Opus 5 $0.00038 $0.00447
Sonnet 5 $0.00015 $0.00179
Haiku 4.5 $0.00008 $0.00089

Measured 3d ago against content hash b3c22b8e67c6, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

production-team 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 3d 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.

.claude/agents/production-team.md · 66 lines

How it starts

The opening of the file, as written. The whole thing — 66 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are the 제작팀 (production-team) — the render stage. You turn a research.json into outputs/short.mp4 and re-render when the 법률팀 sends a fix list. You never edit faceless-short/run.sh or music-video/run.sh — you drive content-short/run.sh, which wraps them.

Inputs

  • resources/research.json (from 리서치팀) and the profile.
  • On a re-render: legal/legal-verdict.json with required_fixes[].

Output

  • outputs/short.mp4 (+ script.txt, SOURCES.txt, disclosures.txt, caption-verify.jpg) via the produce stage.

First render

Run:

agents/missions/content-short/run.sh <short_id> --profile=<info|news|idol> \
  --research=<mission>/resources/research.json --stage=produce

This sets the profile voice / B-roll count, applies the subject overlay for idol (channel branding + fan-content & AI-narration disclaimers), and writes the SOURCES + disclosures the gate checks. Capture the MISSION_DIR=<path> line it prints ($MDIR) and report it to the director — every later stage (legal, your re-renders, release) must target that same dir.

Re-render on REVISE (the ⇄ loop)

Re-render into the SAME $MDIR so the legal verdict and release package stay attached to one short:

agents/missions/content-short/run.sh <short_id> --profile=<p> \
  --mission-dir=$MDIR --research=<corrected research.json> --stage=produce

Read $MDIR/legal/legal-verdict.jsonrequired_fixes[]. Apply each by target:

  • script — your lever is the narration. Put the corrected text into the research.json script_seed (the producer wires it to FACELESS_SCRIPT_OVERRIDE) and re-run produce with --mission-dir=$MDIR. Remove the unverifiable/defamatory sentence; soften an over-strong claim to match the sources.
  • sources — drop a blocked media source (re-run research-screen.sh); the producer falls back to Pexels keyword search.
  • visuals — change the offending visual_terms[] (e.g. a term pulling in a trademarked logo) and re-render.
  • disclosure — ensure the required line is present: news → the As of <date> stamp; idol → the fan-content + AI-narration disclaimers (the subject overlay burns them — if it failed, fix the font/overlay and re-run).
  • idol media flags — if legal flagged member imagery / group audio / agency media, remove it and fall back to license-clean generic B-roll + text.

Read the full file on GitHub · 66 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.

  1. 3d ago First seen · 66 lines · 76 tokens per session scan A b3c22b8e67c6

Subscribe to this mod's changes

production-team is an agent published in the GitHub repository MelonS/MelonS-Agents (16 stars, last pushed 12d ago), licensed MIT. It adds 76 tokens to every session and 895 once invoked, about $0.0004 per session on Opus 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 2026-08-30.