Cc-skills upgrade

Upgrade Kokoro TTS engine dependencies and model. TRIGGERS - upgrade kokoro, update tts, kokoro update, update mlx-audio.

install

source · Clone the upstream repo

git clone https://github.com/terrylica/cc-skills

Claude Code · Install into ~/.claude/skills/

T=$(mktemp -d) && git clone --depth=1 https://github.com/terrylica/cc-skills "$T" && mkdir -p ~/.claude/skills && cp -r "$T/plugins/kokoro-tts/skills/upgrade" ~/.claude/skills/terrylica-cc-skills-upgrade && rm -rf "$T"

manifest: plugins/kokoro-tts/skills/upgrade/SKILL.md

source content

Upgrade Kokoro TTS

Upgrade MLX-Audio dependencies, re-download the model, and update bundled scripts.

Self-Evolving Skill: This skill improves through use. If instructions are wrong, parameters drifted, or a workaround was needed — fix this file immediately, don't defer. Only update for real, reproducible issues.

Workflow

Step 1: Pre-upgrade health check

PLUGIN_DIR="${CLAUDE_PLUGIN_ROOT:-$HOME/.claude/plugins/marketplaces/cc-skills/plugins/kokoro-tts}"
bash "$PLUGIN_DIR/scripts/kokoro-install.sh" --health
cat ~/.local/share/kokoro/version.json

Step 2: Execute upgrade

PLUGIN_DIR="${CLAUDE_PLUGIN_ROOT:-$HOME/.claude/plugins/marketplaces/cc-skills/plugins/kokoro-tts}"
bash "$PLUGIN_DIR/scripts/kokoro-install.sh" --upgrade

This upgrades:

Python packages:
```
mlx-audio
```
,
```
soundfile
```
,
```
numpy
```
Model weights: re-downloaded from
```
mlx-community/Kokoro-82M-bf16
```
Bundled scripts:
```
kokoro_common.py
```
and
```
tts_generate.py
```
re-copied from plugin
```
version.json
```
: rewritten with new versions

Step 3: Post-upgrade verification

PLUGIN_DIR="${CLAUDE_PLUGIN_ROOT:-$HOME/.claude/plugins/marketplaces/cc-skills/plugins/kokoro-tts}"
bash "$PLUGIN_DIR/scripts/kokoro-install.sh" --health
cat ~/.local/share/kokoro/version.json

# Test synthesis
~/.local/share/kokoro/.venv/bin/python ~/.local/share/kokoro/tts_generate.py \
  --text "Upgrade verification" --voice af_heart --lang en-us --speed 1.0 \
  --output /tmp/kokoro-upgrade-test.wav && echo "OK"

Rollback

If upgrade breaks TTS, do a clean reinstall:

PLUGIN_DIR="${CLAUDE_PLUGIN_ROOT:-$HOME/.claude/plugins/marketplaces/cc-skills/plugins/kokoro-tts}"
bash "$PLUGIN_DIR/scripts/kokoro-install.sh" --uninstall
bash "$PLUGIN_DIR/scripts/kokoro-install.sh" --install

Model cache is preserved across uninstall, so reinstall reuses the cached model.

Troubleshooting

Issue Cause Solution

Upgrade fails No internet or PyPI down Check connectivity, retry

Import error after

mlx-audio incompatibility

Clean reinstall:

--uninstall

then

--install

Model download slow Large download Wait for HuggingFace download to complete

Post-Execution Reflection

After this skill completes, reflect before closing the task:

Locate yourself. — Find this SKILL.md's canonical path before editing.
What failed? — Fix the instruction that caused it.
What worked better than expected? — Promote to recommended practice.
What drifted? — Fix any script, reference, or dependency that no longer matches reality.
Log it. — Evolution-log entry with trigger, fix, and evidence.

Do NOT defer. The next invocation inherits whatever you leave behind.