| name | talk-birgitta-closing-keynote |
| description | Answers questions about, retrieves verbatim quotes from, explains concepts from, and summarizes key arguments in Birgitta Böckeler's talk "State of Play: AI Coding Assistants" (AI Native Dev conference, 2026). Use when the user asks about the last 12 months in AI coding assistants, the Opus 4.5 moment, LLM statelessness, context window and attention trade-offs, choosing the right model for a task, the ecosystem around models, or her Thoughtworks/Martin Fowler-site writing on AI-assisted software delivery. |
| metadata | {"generated-by":"talk-to-skill","source":"user-pasted-transcript","generated-at":"2026-06-04T00:00:00.000Z"} |
State of Play: AI Coding Assistants — Birgitta Böckeler
Grounding rules — MUST follow when answering
- Before answering any specific question, read
outline.md to locate the relevant section, then read that section of transcript.md.
- When attributing words, quote verbatim from
transcript.md. Never put quotation marks around paraphrased content.
- If a claim isn't in
transcript.md, say so explicitly — do not fabricate or present paraphrased content as a direct quote. It is fine to say "She doesn't appear to address this in the talk."
- When a question spans multiple sections, read each relevant section of
transcript.md before answering, and note when the speaker addresses the topic in more than one place.
- When a user asks about something adjacent to but not directly covered in the talk, answer from the transcript as far as it goes, then clearly signal where the transcript ends and any broader context begins.
Expected output format
For most questions, structure your response as:
- A brief direct answer or summary in your own words.
- One or more supporting verbatim quotes from
transcript.md, introduced with a neutral phrase such as "In her words:" or "She puts it this way:".
- If relevant, a note on where in the talk this appears (section name from
outline.md).
Example
User: What does Böckeler mean by the "Opus 4.5 moment"?
Response: She uses "the Opus 4.5 moment" to mark a qualitative leap in model capability that changed what teams could realistically delegate to an AI coding assistant. In her words: [verbatim quote from transcript.md]. This comes up in the [section name] section of the talk.
Edge-case guidance
- Question not covered: If the topic is not addressed anywhere in
transcript.md, say so clearly — e.g., "This doesn't appear to be covered in the talk." Do not speculate or fill the gap with outside knowledge unless the user explicitly asks for broader context beyond the talk.
- Partial or inexact coverage: If the transcript touches on a topic but doesn't fully answer the question, or if you cannot locate an exact verbatim quote matching what the user is looking for, share what the transcript does say with the closest relevant passage, label it clearly (e.g., "near match" or "closest passage I can find"), and note the gap — e.g., "The talk touches on this briefly but doesn't go into further detail."
- Ambiguous question: If a question could match multiple sections of
outline.md, check each candidate section in transcript.md before answering, and note which parts of the talk are relevant.
- Questions about her other writing: If the user asks about Böckeler's articles on the Thoughtworks or Martin Fowler site, answer only from content present in
transcript.md or the bundle. Do not retrieve or fabricate content from external URLs.
Key quotes
quotes.md contains pre-extracted verbatim highlights from this talk, organised by theme. When formulating answers, check quotes.md first for strong citable evidence before searching the full transcript.md.
