What is tunnel-policy?
tunnel-policy is a Claude Code mod by ucsandman that mIDDLEWARE TUNNEL stage 3: Policy: allows everything harmless; denies a matching rule. It is listed under Security and has 22 GitHub stars.
| Type | Claude code mod |
|---|---|
| Author | ucsandman |
| Repository | ucsandman/Agnostic-AI |
| Popularity | 22 GitHub stars |
| Category | Security |
| Security | Low risk (96/100) |
| Rank | #308 of 5,272 Claude Code mods; #53 in Security |
| License | MIT |
| Last commit | 2026-10-02 |
| Hooks | 1 events |
| Sees | Bash calls |
| Draws on | nothing |
Is tunnel-policy safe to install?
tunnel-policy scores 96/100 on the MCPMods security check: Low risk. No checks raised a concern. It passed 7 of 8 checks.
Mod footprints come from claude plugin validate, which lists every event a mod hooks and every API call it makes without running it.
| Check | Result |
|---|---|
| No network access | Pass |
| Starts no processes | Pass |
| Writes no files or config | Pass |
| Doesn't read environment variables | Pass |
| Only sees the events it hooks | Pass |
| Doesn't drive Claude | Pass |
| Validator reported problems | Note |
| Has an open-source license | Pass |
API calls in its code
| Call | What it means |
|---|---|
$.ui.log | draws interface |
How do I install tunnel-policy?
- Check your Claude Code version. Mods need Claude Code 2.1.287 or later and are on by default.
claude --version
- Read its footprint. This lists every event it hooks and every API call it makes, without running it.
git clone https://github.com/ucsandman/Agnostic-AI claude plugin validate ./Agnostic-AI/labs/claude-mods/lab/demos/tunnel/policy
- Install it. Run this in your terminal, then /reload-plugins in an open session.
git clone https://github.com/ucsandman/Agnostic-AI && claude --plugin-dir ./Agnostic-AI/labs/claude-mods/lab/demos/tunnel/policy
Or ask Claude to set it up
Paste this into Claude Code and it installs tunnel-policy for you:
Install the "tunnel-policy" Claude Code mod from https://github.com/ucsandman/Agnostic-AI/tree/master/labs/claude-mods/lab/demos/tunnel/policy. Run claude plugin validate on it first and show me what it can access before you enable it.
More from ucsandman/Agnostic-AI
This repository has at least 9 Claude Code mods on MCPMods.
- claude-runtime: Shared Claude runtime adapter: normalises function-hook events into ClaudeRuntimeEvent, exposes $.runtime (emit bus, sup (High risk)
- harness-mods: Harness Mod layer: heartbeat/canary, native usage, rewrite explanations, measured subagent accounting, supervisor routin (Review before installing)
- bender: REALITY BENDER: intercepts a harmless tool result and alters what Claude receives; prints REAL vs RECEIVED (Low risk)
- blackbox: BLACK BOX RECORDER: structured per-session timeline of prompts, model steps, tool calls, subagents and denials; /blackbo (Review before installing)
- guardian: GUARDIAN: executable policy that protects lab/demos/protected/** from Write/Edit/Bash, whatever CLAUDE.md says (Low risk)
- matrix: MATRIX: freezes a harmless tool call before it runs, shows what Claude intended, asks you, then continues or denies. /ma (Low risk)
- tunnel-cost: MIDDLEWARE TUNNEL stage 2: Cost analyzer: estimates the call's token cost from the args size (Low risk)
- tunnel-observer: MIDDLEWARE TUNNEL stage 1: Observer: sees the call first, stamps t0, prints what enters (Low risk)