anvilcv-extract

AnvilCV Context Extractor

Read the full extraction protocol at ${CLAUDE_SKILL_DIR}/content_extract.md and follow it exactly.

That file tells you:

• which two questions to ask the developer before exploring,
• which git commands to run for ownership / contributor checks,
• the exact headed-section output format to emit.

Do not begin exploring the codebase or writing any section until the developer has answered the opening questions, as instructed in the protocol file.

Two hard rules from the protocol, restated so they are not missed:

• Ask the three opening questions first, then STOP and wait for the reply before doing anything else.
• The final output is a file, not chat. Write the completed document to anvilcv-context.md in the repo root (current working directory), overwriting if present. In chat, report only the file path plus a 2–3 line summary — never paste the section contents.