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 kalyvask/winning-writing --skill speech-coachgit clone --depth 1 https://github.com/kalyvask/winning-writingWrote 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/kalyvask/winning-writing/speech-coach)<a href="https://agentmods.dev/skills/kalyvask/winning-writing/speech-coach"><img src="https://agentmods.dev/badge/skills/kalyvask/winning-writing/speech-coach/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/kalyvask/winning-writing/speech-coach"><img src="https://agentmods.dev/badge/skills/kalyvask/winning-writing/speech-coach.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 | $0.00168 | $0.02253 |
| Opus 5 | $0.00084 | $0.01126 |
| Sonnet 5 | $0.00034 | $0.00451 |
| Haiku 4.5 | $0.00017 | $0.00225 |
Grade A, and why
speech-coach 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 10d 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 — 117 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Speech coach
Source: points/speech-rules.md (the 12 rules + the order-of-operations drafting flow + the Sparkline), points/delivery-rules.md (the three Vs, derailers, Q&A structures — for Mode 3 coaching), points/kramon-master.md (rhythm, jargon, warmth), points/banned-jargon.md (especially for spoken delivery — every banned phrase hits even harder when read aloud).
What this skill does
Given a topic + audience (or an existing draft), produce a talk shaped for the ear, not the page. The most common bad-talk failure mode is a memo read aloud — this skill catches that on sight and rebuilds the piece in spoken form.
The philosophy in one line
A speech is not an essay. The reader of an essay can reread a sentence; the listener of a speech cannot. So the closing line matters more than the opening, the main thought must be repeated, and every sentence has to parse on first hearing.
Mode 1 — Draft from scratch
If the user gives you a topic + audience, walk through Kramon's drafting order before writing a single body paragraph:
- Write the closing line first. One sentence. The line you want quoted. If you cannot write it, the user does not have a talk yet — ask them what they want the audience to remember tomorrow morning.
- Write the main thought. One sentence. Different from the closing — the argument, not the image. Said early, repeated once mid-talk.
- Pick ONE story. Real people, real names, one scene. If the user offers three stories, pick a lane — the other two are different talks.
- Pick the opener. Three archetypes work: self-deprecating story, show-of-hands question, single concrete image. Pick the one that earns the laugh or the raised hand faster for this audience. Never "Thank you for having me" — throat-clearing.
- Outline the arc. Opener → main thought → story → second beat or quote → restate main thought → closing line.
- Draft the body in short paragraphs, the way the user actually speaks. Contractions. Dropped subjects where natural. Variable sentence length.
- Cut by one third. Every talk runs long live. Pre-cut so it still fits.
- Read aloud at performance pace with a timer. Mark stumbles — those are listener confusion points; rewrite.
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.
- 10d ago First seen · 117 lines · 168 tokens per session scan A 7b03ad9581a4
speech-coach is a skill published in the GitHub repository kalyvask/winning-writing (13 stars, last pushed 4d ago), licensed MIT. It adds 168 tokens to every session and 2,253 once invoked, about $0.0008 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.
Other skills, from other repositories
project-memory
Generate a project-specific context file from a brief so an AI assistant remembers your editorial constraints, voice, audience, and quality bar across sessions.
project-retrospective
Generate a LESSONS.md from a finished project: what worked, what didn't, what to reuse, what to retire — formatted for next-project carry-over.
template-selector
Recommend the right skill bundle, agent, and workflow sequence for a new project — so media professionals can start producing instead of browsing a 394-skill library.
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.
multi-author-harmonizer
Reviews a text written or assembled by multiple authors and produces a detailed inconsistency report — flagging voice shifts, terminology mismatches, tonal clashes, and formatting discrepancies — with specific harmonisation recommendations for each.
passive-voice-checker
Identifies every passive-voice construction in a draft, assesses whether each weakens or serves the prose, and offers active-voice alternatives for those that should be changed — while leaving justified passives alone.