scriptwriter-draft

A command for creating a shared writers’ room from a starting prompt, with optional first-scene contributions. Participants can exchange signed writing contributions and receive a live event stream.

In plain words
What is it for?
Use it to start a room for scenes or dialogue, view its room and stream details, add the first contribution, or request a random prompt when no seed is provided.
Why use it?
It gives collaborators a defined place and starting idea for building a story together.

Command 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 commands/cambridgetcg/agenttool/scriptwriter-draft
Clone the repo
git clone --depth 1 https://github.com/cambridgetcg/agenttool

Made for: Claude Code.

Per session 35 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 500 The whole file, excluding the scripts and references it only reads on demand.
Security scan A 1 finding. 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 $0.00035 $0.00500
Opus 5 $0.00017 $0.00250
Sonnet 5 $0.00007 $0.00100
Haiku 4.5 $0.00003 $0.00050

Measured yesterday against content hash 83b6a71aa129, method: parsed. Prices are Anthropic first-party input rates as of 2026-08-30, from the pricing page.

Security

Grade A, and why

scriptwriter-draft 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 yesterday.

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 -sS -X POST http://localhost:7777/rooms \
packages/scriptwriter/.claude-plugin/commands/scriptwriter-draft.md · 39 lines

What it actually says

The user wants to start a new writers' room.

$ARGUMENTS is the seed prompt — a starting line every contribution riffs on. If the user gave nothing, ask them once for a seed (4+ characters), or offer to draw a chaos card first to seed the room (mcp__scriptwriter__draw_chaos_card).

Path A — MCP tools available

  1. Call mcp__scriptwriter__create_room with { seed: "$ARGUMENTS" }. The room gets an auto-generated meme-name unless the user specified one.
  2. Echo the room ID, the auto-generated name, the vibe, and the URLs (stream + contribute).
  3. Offer to add the first contribution: ask whether the user wants to seed a scene, dialogue, or draw a chaos card for inspiration.
  4. If they want to contribute, call mcp__scriptwriter__contribute_to_room with the appropriate kind.

Path B — CLI

If MCP isn't configured, the node must be running (bun packages/scriptwriter/bin/scriptwriter.ts serve --port 7777). Then:

curl -sS -X POST http://localhost:7777/rooms \
  -H 'content-type: application/json' \
  -d "$(jq -nc --arg seed "$ARGUMENTS" '{seed: $seed}')"

Report back the returned room ID and URLs.

After creating

Surface the immediate-next moves:

Room <name> (<id>) created. Three things you can do right now:

  1. Draw a chaos card to seed inspiration (/scriptwriter-draw)
  2. Contribute the first scene via mcp__scriptwriter__contribute_to_room or curl POST /rooms/<id>/contributions
  3. Share the room URL with a peer at SYNCED depth in your cascade — they get implicit allowlist

If the room is created free-flow (no allowlist_dids), anyone with a valid signature can contribute. To restrict, re-create with an explicit allowlist of DIDs, or just trust the free-flow default for an open jam.

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. yesterday First seen · 39 lines · 35 tokens per session scan A 83b6a71aa129

Subscribe to this mod's changes

scriptwriter-draft is a command published in the GitHub repository cambridgetcg/agenttool (0 stars, last pushed yesterday), licensed Apache-2.0. It adds 35 tokens to every session and 500 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 2026-08-31.