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 instructions/sandraschi/tailscale-mcp/agents-mdgit clone --depth 1 https://github.com/sandraschi/tailscale-mcpWrote this? Show the measurements
A badge with what this costs and how it scanned, read live from this page, so it follows the numbers instead of freezing them. Markdown for a README, HTML for a documentation site or a project page.
[](https://agentmods.dev/instructions/sandraschi/tailscale-mcp/agents-md)<a href="https://agentmods.dev/instructions/sandraschi/tailscale-mcp/agents-md"><img src="https://agentmods.dev/badge/instructions/sandraschi/tailscale-mcp/agents-md.svg" alt="Measured on agentmods" height="20"></a>What 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 | $0.00776 | $0.00776 |
| Opus 5 | $0.00388 | $0.00388 |
| Sonnet 5 | $0.00155 | $0.00155 |
| Haiku 4.5 | $0.00078 | $0.00078 |
Grade C, and why
tailscale-mcp AGENTS.md 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 3d 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.
or installable via `curl -fsSL https://lmstudio.ai/install.sh | bash`). Makes network callslowCapability
Not a fault in itself. Listed so you know the mod talks to something, and to what.
or installable via `curl -fsSL https://lmstudio.ai/install.sh | bash`). How it starts
The opening of the file, as written. The whole thing — 63 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Agent Workflow Recollection for tailscale-mcp
Project Identity
- Repo: tailscale-mcp — FastMCP 3.2+ Tailscale network controller
- Python: >=3.12, uses
uvfor package management - Framework: FastMCP 3.2 with portmanteau tool pattern
- Webapp: React/Vite frontend (port 10820) + FastAPI backend (port 10821)
Key Architecture
src/tailscalemcp/
tools/ # MCP tool surface (portmanteau pattern)
operations/ # Business logic layer
client/ # Tailscale API client with rate limiting + retry
models/ # Pydantic models
server.py # FastAPI backend for web_sota
mcp_server.py # FastMCP server setup
Common Tasks
- Run all checks:
uv run ruff check . && uv run mypy src/tailscalemcp && uv run pytest - Start webapp:
.\start.ps1(launches backend + frontend) - Start MCP only:
uv run python -m tailscalemcp - Run tests:
uv run pytest -v - Add dependency: Add to
pyproject.tomlthenuv lock && uv sync
Ports Used
- Webapp frontend: 10820
- Backend/MCP HTTP: 10821
- Prometheus metrics: 9091
- Monitoring stack: Prometheus 9090, Loki 3100, Grafana 3000 (all 127.0.0.1 only)
Fleet Standard Alignment
- FastMCP 3.2+ required
- Transport via
src/tailscalemcp/transport.py(stdio/http/sse) - All ports in 10700+ range
- structlog for structured logging (JSON, stderr)
- Prometheus metrics on separate port
LM Link Integration (Tailscale + LM Studio)
LM Link is a Tailscale-powered encrypted mesh for remote LLM access (Feb 2026).
The get_lm_link tool provides operational control over LM Link via the lms CLI:
| Operation | CLI equivalent | Description |
|---|---|---|
status |
lms link status --json |
Live peers, loaded models, link state |
enable |
lms link enable |
Enable LM Link on this device |
disable |
lms link disable |
Disable LM Link |
set_device_name |
lms link set-device-name <name> |
Rename this device |
set_preferred_device |
lms link set-preferred-device <device> |
Set preferred peer |
info |
(static) | Setup docs and links |
readiness |
(Tailscale API) | Tailnet connectivity check |
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.
- 3d ago First seen · 63 lines · 776 tokens per session scan C 8738dad72cda
tailscale-mcp AGENTS.md is an instructions file published in the GitHub repository sandraschi/tailscale-mcp (4 stars, last pushed 18d ago), licensed MIT. It adds 776 tokens to every session, about $0.0039 per session on Opus 5. 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 instructions, from other repositories
blender-mcp AGENTS.md
Instructions for sandraschi/blender-mcp, covering blender-mcp agent context and quick ref.
blender-mcp CLAUDE.md
Instructions for sandraschi/blender-mcp, covering blender-mcp — ai-powered blender automation, commands, architecture and code style & standards.
blender-mcp copilot-instructions.md
Instructions for sandraschi/blender-mcp: You have access to a Blender 3D automation server. You can control Blender scenes, objects, materials, rendering, and physics through MCP tools.
plexmcp AGENTS.md
AGENTS.md instructions for sandraschi/plexmcp, covering plexmcp — agent context, project overview, key commands, lint and test.
streamfog-mcp AGENTS.md
AGENTS.md instructions for sandraschi/streamfog-mcp, covering agent instructions — streamfog-mcp, identity, commands, ports (fleet-registered) and architecture.
discord-mcp AGENTS.md
Instructions for sandraschi/discord-mcp, covering discord-mcp — agent guide, standards, key files and quick ref.