Your Speaker Context Block
A fill-in template you paste at the top of every AI chat so it always knows who you are, who you speak to, and how you sound.
What this is for
Every AI chat starts from zero. It doesn't remember your last conversation, doesn't know your speaking topic, and doesn't know your voice unless you tell it. The Speaker Context Block fixes that. It's one block of text you fill in once and paste at the top of every new chat about your talk.
When to use it
Paste your context block at the start of every session where you're working on a talk - a new idea, a rewrite, a rehearsal, a CFP answer. Update it as your material changes: new proof points, a new core message, a new audience you're targeting.
The prompt kit
Fill in the brackets once and save this somewhere you can copy from fast - a note on your phone, a doc, or the Battle Book in this app.
SPEAKER CONTEXT BLOCK
Who I am: [your name, your role, your one-line bio - the version you'd give a stranger in an elevator]
Who I speak to: [the audience type - e.g. "mid-market HR directors," "regional bank branch managers," "association executives"]
My core message: [the one idea your talks build toward, in one sentence]
My proof points: [2-4 stories, numbers, or credentials you actually have - not aspirational ones]
My voice rules:
- [a phrase or word you always use]
- [a phrase or word you never use]
- [how formal or casual you are on stage]
Never invent facts, statistics, or stories on my behalf. If you need a detail I haven't given you, ask.
Use this second prompt when you want the AI to check its own understanding before you go further:
Here's my Speaker Context Block: [paste it]
Before we start working, tell me back in your own words: who I speak to, what my core message is, and one thing about my voice you'll watch for. If anything is unclear, ask me now instead of guessing later.
What good output looks like
The AI repeats your audience and your message back accurately, and names something specific about your voice - not "professional and engaging," but something you actually wrote, like "you never use the word 'leverage.'" If it paraphrases your core message into something broader or vaguer than what you wrote, correct it before moving on. A vague read of your message now means vague output all session.
Watch out for
Letting the block go stale. Your proof points change as you deliver more talks and collect more results. A context block from six months ago will steer the AI toward material you've already retired. Update it the same day you retire a story or land a new one.
Padding your proof points. This block is not the place to round up. If a result was close to 20 percent, write that, not "20 percent." The AI treats whatever you write here as fact and builds on it. A stretched number here becomes a false claim in your draft.
Making the voice rules too vague to use. "Casual but professional" tells the AI nothing it can act on. "I use contractions, short sentences, and I never say 'leverage' as a verb" gives it something to check against.
Skipping this step because it feels slow. It takes five minutes once. Every session after that starts faster and closer to your voice, because you're not re-explaining who you are every time.