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 JayRHa/AgentSkills --skill public-speaking-coachgit clone --depth 1 https://github.com/JayRHa/AgentSkillsWrote 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/jayrha/agentskills/public-speaking-coach)<a href="https://agentmods.dev/skills/jayrha/agentskills/public-speaking-coach"><img src="https://agentmods.dev/badge/skills/jayrha/agentskills/public-speaking-coach.svg" alt="Measured on agentmods" 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 | $0.00106 | $0.01227 |
| Opus 5 | $0.00053 | $0.00613 |
| Sonnet 5 | $0.00021 | $0.00245 |
| Haiku 4.5 | $0.00011 | $0.00123 |
Grade A, and why
public-speaking-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 8d 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 — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Public Speaking Coach
Overview
This skill helps a speaker prepare and deliver: it distills a single core message, structures the talk around a narrative arc, crafts a memorable opening and closing, tightens the language for the ear (not the page), and coaches delivery, nerves, and Q&A. Works for keynotes, work presentations, pitches, toasts, weddings, and eulogies.
Keywords: public speaking, speech, presentation, talk, keynote, toast, wedding speech, eulogy, pitch, stage fright, speaking anxiety, storytelling, delivery, Q&A, slides.
When to use vs. not
Use this to plan, write, rehearse, or get feedback on any spoken delivery, and to manage speaking anxiety. It coaches structure and delivery — it can't rehearse your body language for you, so it always pushes the speaker to practice out loud. For severe, clinical anxiety, suggest professional support alongside the practical techniques.
Inputs to gather first
- Occasion + audience (who, how many, how formal, what they care about).
- Goal — what should the audience think, feel, or do afterward?
- Time limit and format (slides? podium? virtual? toast at a table?).
- Raw material — the speaker's points, stories, data, or a draft.
- Concerns — nerves, time overrun, a tough crowd, or Q&A.
Workflow
- Find the one core message. Force it into a single sentence the audience should remember. Everything serves this; if it doesn't, cut it. A talk about everything lands nothing.
- Define the goal as an outcome. Inform, persuade, inspire, or entertain — and what the audience does next. This sets tone and structure.
- Choose a structure / arc. Match the occasion: problem→solution, story arc, "What? So what? Now what?", or for toasts the relationship→anecdote→wish shape. One idea per section. See
references/structure-and-stories.md. - Write a strong open and close. Open with a hook (story, question, surprising fact, vivid image) — never "Hi, today I'm going to talk about." Close with a callback to the core message and a clear final line; for persuasion, an explicit call to action. The first and last 30 seconds carry the most weight.
- Write for the ear. Short sentences, concrete words, signposting ("three things…"), pauses, and stories over abstraction. Read it aloud and cut anything that trips the tongue. See
references/delivery-and-nerves.md. - Design slides as support, not script (if any): one idea per slide, visuals over bullet walls, never read them verbatim. The speaker is the message; slides are the echo.
- Rehearse out loud + time it. Practice standing, aloud, ideally recorded. Trim to ~90% of the limit. Internalize beats and transitions, not a word-for-word script (which sounds robotic and breaks under nerves).
- Prep nerves + Q&A. A breathing/warm-up routine, a reframe of adrenaline as energy, anticipated questions with crisp answers, and a graceful "I don't know — I'll follow up" line. See
references/delivery-and-nerves.md.
What ships with it
3 files beside SKILL.md in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.
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.
- 8d ago First seen · 63 lines · 106 tokens per session scan A de2f27111460
public-speaking-coach is a skill published in the GitHub repository JayRHa/AgentSkills (5 stars, last pushed 1mo ago), licensed MIT. It adds 106 tokens to every session and 1,227 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
code-explainer
Paste any code snippet and get a plain English explanation of what it does and how it works.
student-learning
Use when the user wants to learn or study a topic, understand a concept, prepare for an exam, or be tutored. Teaches with Socratic questioning and teach-back so the user actually learns — it does not just hand over answers to copy.
canvas
Fetch Canvas LMS courses and assignments via API token.
rwkv-architecture
RNN+Transformer hybrid with O(n) inference. Linear time, infinite context, no KV cache. Train like GPT (parallel), infer like RNN (sequential). Linux Foundation AI project. Production at Windows, Office, NeMo. RWKV-7 (March 2025). Models up to 14B parameters.
stride-analysis-patterns
Apply STRIDE methodology to systematically identify threats. Use when analyzing system security, conducting threat modeling sessions, or creating security documentation.
agent-patterns-catalog
A reference catalogue of six common AI-agent designs, including agents that reason and act, retrieve documents, call tools, coordinate several agents, or work autonomously. It compares their structures, data flow, and suitable uses without tying them to one framework.