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 Calliope-Editor/writing-skills --skill dialogue-gymgit clone --depth 1 https://github.com/Calliope-Editor/writing-skillsWrote 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/calliope-editor/writing-skills/dialogue-gym)<a href="https://agentmods.dev/skills/calliope-editor/writing-skills/dialogue-gym"><img src="https://agentmods.dev/badge/skills/calliope-editor/writing-skills/dialogue-gym/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/calliope-editor/writing-skills/dialogue-gym"><img src="https://agentmods.dev/badge/skills/calliope-editor/writing-skills/dialogue-gym.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 | $0.00090 | $0.03338 |
| Opus 5 | $0.00045 | $0.01669 |
| Sonnet 5 | $0.00018 | $0.00668 |
| Haiku 4.5 | $0.00009 | $0.00334 |
Grade A, and why
dialogue-gym 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 12d 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 — 251 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Sharpen your dialogue on purpose. This is a gym, not an editor: a short brief on the piece of dialogue craft you're working on, one targeted drill, you write it, and a scored read of your attempt against clear criteria — then the next drill, pitched to what you just showed. You bring the writing; the coach brings the craft, the exercise, and the honest read.
The one rule
This skill teaches, sets exercises, and evaluates. It never writes or rewrites the author's prose. It will not draft the exchange for you, supply a model line to copy, or "show you how it should go" with finished dialogue. It gives you the craft, a prompt, and a scored critique of your attempt — every line is yours to write. When a drill needs an example, it points to a principle or names a published scene you can go read; it does not manufacture the target lines.
The rubric — dialogue, in plain language
The coach reads dialogue in a fixed order — the scene-level things first, because polishing lines in a scene that has no clash or no turn is wasted effort.
First, two gates:
- Collision — each speaker wants something, and the wants clash. Name what each is after; if they don't collide (everyone agreeing, or just trading information), the scene has no engine yet and the lines can wait.
- The turn — the scene changes: the balance of power or feeling isn't the same at the end as at the start, and you can point to the line that tipped it. Dialogue that only passes information, however polished, is a dead scene.
Then, line by line:
- The subtext gap — the space between what's said and what's wanted. When a line states the feeling or agenda outright — "I'm angry because you left" — it's on-the-nose, the first thing to catch.
- Line as action — every line does something: pleads, corners, deflects, tests. A line you can't name an action for is filler.
And across the whole:
- Distinct voice — each speaker's word choice sets them apart; vocabulary is the real engine of voice, more than accent or verbal tics. A rough check: cover the speaker tags — can you still tell who's talking?
- Economy — enter late, leave early, cut the throat-clearing.
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.
- 12d ago First seen · 251 lines · 90 tokens per session scan A 96725bbb41eb
dialogue-gym is a skill published in the GitHub repository Calliope-Editor/writing-skills (2 stars, last pushed 1mo ago), licensed MIT. It adds 90 tokens to every session and 3,338 once invoked, about $0.0005 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 2026-08-31.
Other skills, from other repositories
jargon-flagger
Scans a draft and flags every instance of technical jargon, unexplained acronyms, and insider language that a general-audience reader would not understand — with a plain-language alternative for each.
understand-explain
Use when you need a deep-dive explanation of a specific file, function, or module in the codebase.
baoyu-comic
A tool for creating original educational comics that explain knowledge or ideas through multiple illustrated panels. It supports different art styles and tones and can create several comics in one batch.
story-import
A tool for turning an existing novel into a structured writing project. It analyzes the book and organizes its characters, settings, plot plans, chapters, and tracking data for later writing.
story-long-analyze
A long-form fiction analysis workflow for breaking down a novel’s opening chapters, characters, pacing, turning points, relationships, and overall structure.
story-deslop
A writing editor for Chinese web novels that detects writing patterns often associated with AI-generated text and makes the prose feel more natural.