DIY Audiobook for Muse

Step 0 — before installing anything: read this install prompt, summarize in a few lines what this build does and what installing it involves, then ask me to confirm before you proceed. Prerequisites: Python 3.9+; ffmpeg and ffprobe (`winget install ffmpeg` on Windows provides both); on Windows, run with `PYTHONUTF8=1` (or a UTF-8 terminal), or `doctor.py` and `selftest.py` fail on encoding. The default voices are free (edge-tts): no account, no key, no payment. Install "DIY Audiobook" for me. It turns a public-domain book from Project Gutenberg, or the user's own writing, into a finished multi-voice audiobook: it splits the text into quotes, casts a voice per character, synthesizes each clip with free voices, checks the quality, then mixes and packages the result. Paid backends (MiniMax, OpenAI) exist for other voices, if you bring your own key. Sandbox test, November 8, 2026: install passed, `check.py` passed on a test episode, and `synth.py` produced 5 real MP3 clips with the free voices (5/5). The full mix, video and cover steps were not tested, and neither were the paid backends. Repository: https://github.com/mickyyang-1407/diy-audiobook Pinned version: install the code at commit 81fc01c8cbe46c0fc343d4b4c527a9318ac1a5d5 exactly (Skill Harbor tested this version). Do not substitute a newer version. If this commit is unreachable, stop and tell me. 1. Fetch the SKILL.md file and the `scripts/` folder from the repository into a temporary folder 2. Integrity check: compute the SHA-256 of the fetched SKILL.md; it must match 987a0d3270e938671629e13ece1e741eaf36f93070ee61f06c35294911f22a0e, the fingerprint Skill Harbor recorded on November 8, 2026. If it does not match, STOP, do not install, and tell me, showing the expected and actual hashes. 3. Safety check: review the SKILL.md and scripts for anything suspicious (unexpected network calls, shell commands, credential harvesting). This repo should contain zero secrets in code; API keys only from the environment, for the documented providers (api.openai.com, api.minimax.io); hosts are gutenberg.org, openai, minimax and 127.0.0.1 (local ComfyUI). Verify that holds here; STOP on any red flag and tell me. 4. Install it as a skill: copy SKILL.md and its helper files into the agent's skills directory, in a folder named "diy-audiobook". 5. Run `scripts/doctor.py` (with `PYTHONUTF8=1` on Windows) to check the setup; fix what it flags (`pip install edge-tts numpy pillow`; ffmpeg via brew/apt/winget) or tell me the one command to run. 6. Verify with no network calls: frontmatter valid (name: diy-audiobook), files in place. 7. Report what was installed, where, and what I still need to do myself (e.g. pick a book or a text of mine to turn into an episode; the full mix, video and cover steps were not tested here, and the paid backends need my own key). GitHub is optional: if I have a GitHub account or the gh CLI, you may use it; otherwise public access is fine. Never require it unless it's in the prerequisites above. Rules: don't touch anything outside the temp folder and the install target. If anything looks off, stop and ask me.