add-llama-cpp
Installs and configures a local llama.cpp server for efficient text generation experiments, integrating with Deus for enhanced performance.
Install this skill
Security score
The add-llama-cpp skill was audited on Jun 1, 2026 and we found 26 security issues across 4 threat categories. Review the findings below before installing.
Categories Tested
Security Issues
Template literal with variable interpolation in command context
| 78 | ```bash |
Curl to non-GitHub URL
| 26 | curl -fsS http://127.0.0.1:8080/health 2>/dev/null || echo "llama-server not responding on 127.0.0.1:8080" |
Curl to non-GitHub URL
| 158 | curl -fsS http://127.0.0.1:8080/health |
Curl to non-GitHub URL
| 159 | curl -fsS http://127.0.0.1:8080/v1/models |
Curl to non-GitHub URL
| 165 | curl -fsS http://127.0.0.1:8080/v1/chat/completions \ |
Access to hidden dotfiles in home directory
| 259 | If the chosen Hugging Face preset is invalid, switch `LLAMA_CPP_MODEL` in `~/.config/deus/llama-cpp.env` to a known-good preset and restart the service. |
Access to .env file
| 67 | cat > "$HOME/.config/deus/llama-cpp.env" <<'EOF' |
Access to .env file
| 83 | ENV_FILE="$HOME/.config/deus/llama-cpp.env" |
Access to .env file
| 201 | if grep -q '^LLAMA_CPP_BASE_URL=' .env 2>/dev/null; then |
Access to .env file
| 202 | tmpf=$(mktemp) && sed 's#^LLAMA_CPP_BASE_URL=.*#LLAMA_CPP_BASE_URL=http://127.0.0.1:8080#' .env > "$tmpf" && mv "$tmpf" .env |
Access to .env file
| 204 | echo 'LLAMA_CPP_BASE_URL=http://127.0.0.1:8080' >> .env |
Access to .env file
| 207 | if grep -q '^LLAMA_CPP_MODEL=' .env 2>/dev/null; then |
Access to .env file
| 208 | tmpf=$(mktemp) && sed 's#^LLAMA_CPP_MODEL=.*#LLAMA_CPP_MODEL=ggml-org/gemma-3-1b-it-GGUF:Q4_K_M#' .env > "$tmpf" && mv "$tmpf" .env |
Access to .env file
| 210 | echo 'LLAMA_CPP_MODEL=ggml-org/gemma-3-1b-it-GGUF:Q4_K_M' >> .env |
Access to .env file
| 213 | mkdir -p data/env && cp .env data/env/env |
Access to .env file
| 259 | If the chosen Hugging Face preset is invalid, switch `LLAMA_CPP_MODEL` in `~/.config/deus/llama-cpp.env` to a known-good preset and restart the service. |
Access to .env file
| 279 | rm -f "$HOME/.config/deus/llama-cpp.env" |
External URL reference
| 26 | curl -fsS http://127.0.0.1:8080/health 2>/dev/null || echo "llama-server not responding on 127.0.0.1:8080" |
External URL reference
| 118 | <!DOCTYPE plist PUBLIC "-//Apple//DTD PLIST 1.0//EN" "http://www.apple.com/DTDs/PropertyList-1.0.dtd"> |
External URL reference
| 158 | curl -fsS http://127.0.0.1:8080/health |
External URL reference
| 159 | curl -fsS http://127.0.0.1:8080/v1/models |
External URL reference
| 165 | curl -fsS http://127.0.0.1:8080/v1/chat/completions \ |
External URL reference
| 202 | tmpf=$(mktemp) && sed 's#^LLAMA_CPP_BASE_URL=.*#LLAMA_CPP_BASE_URL=http://127.0.0.1:8080#' .env > "$tmpf" && mv "$tmpf" .env |
External URL reference
| 204 | echo 'LLAMA_CPP_BASE_URL=http://127.0.0.1:8080' >> .env |
External URL reference
| 225 | LLAMA_CPP_BASE_URL=http://127.0.0.1:8080 npx tsx setup/index.ts --step llama-cpp |
External URL reference
| 233 | LLAMA_CPP_BASE_URL=http://127.0.0.1:8080 python3 -m evolution.benchmark_generative \ |