Getting it into your agent
One page per mod, every tool's command on it. A separate URL per tool would split the same page into five that compete with each other.
npx agentmods add commands/swoopeagle/standardgraph/devicesgit clone --depth 1 https://github.com/swoopeagle/standardgraphWhat it costs to keep this loaded
Counted locally with the o200k_base tokenizer, which is exact for GPT models; Claude uses its own tokenizer and its counts differ. Treat this as one consistent yardstick across the catalogue rather than a bill. Prices are per million input tokens.
| Model | Per session | Once invoked |
|---|---|---|
| Fable 5 | $0.00000 | $0.01105 |
| Opus 5 | $0.00000 | $0.00553 |
| Sonnet 5 | $0.00000 | $0.00221 |
| Haiku 4.5 | $0.00000 | $0.00111 |
Grade C, and why
devices scanned grade C with 2 findings against 26 rules in 11 categories — prompt injection, anti-refusal, data exfiltration, privilege escalation, supply chain, agent snooping, system-prompt leakage, SSRF and excessive agency — measured 2d ago.
A static scan of the body, not an audit. Every finding is printed with the line that produced it so you can judge whether it matters here. A mod is markdown that instructs an agent; that is exactly why what it instructs is worth reading.
Downloads and executes remote codehighSupply chain
curl | sh runs whatever the server returns today, which is not necessarily what it returned when this was reviewed.
curl -sf --max-time 5 http://100.77.63.73:11434/api/ps | python3 -c "import sys,json; d=json.load(sys.stdin); models=d.get('models',[]); print('Ollama UP —', [m['name'] + ' ' + str(round(m.get('size_vram',0)/1e9,1)) + 'G Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
curl -sf --max-time 5 http://100.77.63.73:11434/api/ps | python3 -c "import sys,json; d=json.load(sys.stdin); models=d.get('models',[]); print('Ollama UP —', [m['name'] + ' ' + str(round(m.get('size_vram',0)/1e9,1)) + 'G What it actually says
Check the status of all machines on the StandardGraph Tailscale network.
Tailscale devices:
- Mac Studio: [email protected] (Ollama host, no repo)
- Mac mini 2: [email protected] (pipeline + MCP server)
- Mac mini 3: [email protected] (pipeline + MCP server)
- IWPC: 100.70.170.62 (Windows, SSH not yet set up)
- MacBook Pro (local): 100.118.151.10
Project path on Mac minis: ~/projects/intl-math-standards-mcp/
Step 1 — ping all remotes in parallel:
ping -c 1 -W 2000 100.77.63.73 && echo "Mac Studio REACHABLE" || echo "Mac Studio UNREACHABLE"
ping -c 1 -W 2000 100.101.100.96 && echo "Mac mini 2 REACHABLE" || echo "Mac mini 2 UNREACHABLE"
ping -c 1 -W 2000 100.123.114.101 && echo "Mac mini 3 REACHABLE" || echo "Mac mini 3 UNREACHABLE"
ping -c 1 -W 2000 100.70.170.62 && echo "IWPC REACHABLE" || echo "IWPC UNREACHABLE"
Step 2 — Mac Studio (Ollama only, no SSH needed for status):
curl -sf --max-time 5 http://100.77.63.73:11434/api/ps | python3 -c "import sys,json; d=json.load(sys.stdin); models=d.get('models',[]); print('Ollama UP —', [m['name'] + ' ' + str(round(m.get('size_vram',0)/1e9,1)) + 'GB' for m in models] or ['idle'])"
Step 3 — Mac mini 2:
ssh -o ConnectTimeout=5 [email protected] "
echo '--- uptime ---' && uptime &&
echo '--- ollama ---' && (curl -sf http://localhost:11434/api/ps | python3 -c 'import sys,json; d=json.load(sys.stdin); models=d.get(\"models\",[]); print(len(models),\"model(s):\", [m[\"name\"] for m in models])' 2>/dev/null || echo 'not running') &&
echo '--- pipeline ---' && (ps aux | grep -E 'overnight|reingest|embed|relate|crosswalk' | grep -v grep || echo 'none') &&
echo '--- lock ---' && (ls ~/projects/intl-math-standards-mcp/.pipeline.lock 2>/dev/null && echo 'LOCKED' || echo 'no lock') &&
echo '--- db ---' && sqlite3 ~/projects/intl-math-standards-mcp/data/common_core.db \"SELECT COUNT(*) || ' standards, ' || COUNT(DISTINCT system) || ' systems' FROM standards;\" 2>/dev/null &&
echo '--- disk ---' && df -h / | tail -1
"
Step 4 — Mac mini 3 (same as above with different IP):
ssh -o ConnectTimeout=5 [email protected] "
echo '--- uptime ---' && uptime &&
echo '--- ollama ---' && (curl -sf http://localhost:11434/api/ps | python3 -c 'import sys,json; d=json.load(sys.stdin); models=d.get(\"models\",[]); print(len(models),\"model(s):\", [m[\"name\"] for m in models])' 2>/dev/null || echo 'not running') &&
echo '--- pipeline ---' && (ps aux | grep -E 'overnight|reingest|embed|relate|crosswalk' | grep -v grep || echo 'none') &&
echo '--- lock ---' && (ls ~/projects/intl-math-standards-mcp/.pipeline.lock 2>/dev/null && echo 'LOCKED' || echo 'no lock') &&
echo '--- db ---' && sqlite3 ~/projects/intl-math-standards-mcp/data/common_core.db \"SELECT COUNT(*) || ' standards, ' || COUNT(DISTINCT system) || ' systems' FROM standards;\" 2>/dev/null &&
echo '--- disk ---' && df -h / | tail -1
"
Report as a clean table:
| Device | Status | Ollama | Pipeline | DB | Disk free |
and note any lock files (stale from power loss — safe to remove with rm .pipeline.lock).
If a pipeline was in progress on either Mac mini, tail the latest log:
ssh [email protected] "ls -t ~/projects/intl-math-standards-mcp/logs/*.log | head -1 | xargs tail -20"
What this file has done since we first saw it
Hashed on every crawl. A supply-chain change to an agent config is a question of when, not whether, so the history is kept rather than the latest state alone.
- 2d ago First seen · 57 lines · 0 tokens per session scan C 5dc1b61983a1
devices is a command published in the GitHub repository swoopeagle/standardgraph (5 stars, last pushed 1mo ago), licensed MIT. It costs nothing until one of its globs matches a file; then it loads 1,105 tokens. A static security scan graded it C with 2 findings (downloads and executes remote code, makes network calls). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-31.
Other commands, from other repositories
setup-elevenlabs.es
Command "setup-elevenlabs.es" from minicoohei/ai-agent-camp, covering configuración de la api de elevenlabs, step 0: verificar el progreso de configuración, lo que hará en esta sesión, verificación de preparación and step 1: abrir elevenlabs en el navegador.
setup-elevenlabs
Command "setup-elevenlabs" from minicoohei/ai-agent-camp, covering elevenlabs api セットアップ, step 0: セットアップ進捗の確認, このセッションでやること, 準備チェック and step 1: ブラウザでelevenlabsを開く.
setup-github
Command "setup-github" from minicoohei/ai-agent-camp, covering github アカウント設定とリポジトリ作成, step 0: セットアップ進捗の確認, このセッションでやること, セッション開始前の確認 and step 1: github アカウントの確認.
setup-bigquery.en
Command "setup-bigquery.en" from minicoohei/ai-agent-camp, covering bigquery / gcp authentication setup, step 0: check setup progress, what you'll do in this session, readiness check and step 1: install gcloud cli.
setup-bigquery.es
Command "setup-bigquery.es" from minicoohei/ai-agent-camp, covering configuración de autenticación bigquery / gcp, step 0: verificar el progreso de configuración, lo que hará en esta sesión, verificación de preparación and step 1: instalación de gcloud cli.
setup-bigquery
Command "setup-bigquery" from minicoohei/ai-agent-camp, covering bigquery / gcp 認証セットアップ, step 0: セットアップ進捗の確認, このセッションでやること, 準備チェック and step 1: gcloud cli のインストール.