Stats
Actions
Tags
From ollama
List installed Ollama models with size and modification date. Reads host from ~/.claude/ollama.json.
How this skill is triggered — by the user, by Claude, or both
Slash command
/ollama:ollama-modelsThe summary Claude sees in its skill listing — used to decide when to auto-load this skill
```bash
if [ ! -f "$HOME/.claude/ollama.json" ]; then
echo "ERROR: Ollama not configured. Run /ollama-setup first."
exit 1
fi
HOST=$(python3 -c "import json,os; d=json.load(open(os.path.expanduser('~/.claude/ollama.json'))); print(d.get('host','http://localhost:11434'))")
CURRENT=$(python3 -c "import json,os; d=json.load(open(os.path.expanduser('~/.claude/ollama.json'))); print(d.get('model',''))")
RESULT=$(curl -sf "$HOST/api/tags")
if [ -z "$RESULT" ]; then
echo "Cannot reach Ollama at $HOST — run /ollama-setup"
exit 1
fi
echo "Ollama models at $HOST (active: $CURRENT)"
echo "---"
echo "$RESULT" | python3 -c "
import json, sys
data = json.loads(sys.stdin.read())
models = data.get('models', [])
if not models:
print('No models installed.')
else:
for m in models:
size_gb = round(m.get('size', 0) / 1073741824, 1)
modified = m.get('modified_at', '')[:10]
print(f\"{m['name']:<40} {size_gb}GB {modified}\")
"
Creates, edits, and optimizes skills for Claude Code, including drafting, evaluating with test prompts, iterating on performance, and improving skill descriptions for better triggering accuracy.
npx claudepluginhub danielstewart77/ollama-claude-plugin --plugin ollama