What is Interlinked CLI?

A local guard layer for AI coding agents that evaluates delivered tool hooks against deterministic rules, blocks dangerous ones before they run, and writes a replayable local activity log. It calls itself the harness for your harness.

How does it differ from the agent's own permissions?

It sits at the provider integration layer where an agent's intent becomes a real action, adds deny rules based on content — such as secrets in proposed writes or unvetted installs — and keeps an audit trail the agent's own permission prompt does not.

With Node.js 22+ on macOS, Linux or WSL: npm install -g github:QuentinCody/interlinked-cli, then run interlinked inside the repo you want guarded. The wizard reviews the posture, installs hooks and skills, and starts the daemon.

Does it require a cloud service?

No. The README says the local guard requires no remote service and runtime state lives under .interlinked/. Optional authenticated sync is explicit.

What do the audit numbers mean?

Over 109 days on the author's machine the activity log captured 3,834 guard_block events, and 911 survived a per-event audit against Claude Code session transcripts — including 580 blocked edits that introduced a new TypeScript error.

Does it work with every agent?

Claude Code and Codex are the supported pair; Cursor, Copilot CLI, Gemini CLI, OpenCode and Pi adapters are experimental, and each can enforce only the controls its native API exposes.

How do I turn it off?

interlinked disable stands the guard down non-destructively. interlinked disable --uninstall removes an enable or wizard installation, and uninstall-hooks removes only hooks installed through install-hooks; both preserve a managed bridge if its bytes changed.

Jev directory

CommunityJev Tools & CLIs

Interlinked CLI

A local guard layer for AI coding agents: it evaluates delivered tool hooks against deterministic rules, blocks dangerous ones in milliseconds with no model in the decision path, and keeps a grep-able local activity log for auditing.

See highlights Open github.com

Stars: 180
Language: TypeScript
License: MIT
Reading time: 2 min
Published by: Community
Author: quentincody
Last verified: Oct 2026

quentincody/interlinked-cli

User repository on GitHub

View on GitHub

The harness for your harness. Local hooks, taste enforcement, and developer observability for AI coding agents (Claude Code, Codex, Cursor, Copilot CLI).

Forks: 7
Last pushed: Updated Oct 2026

Repo stats from the GitHub API, cached Oct 2026.Homepage

Tagscommunity catalog-import logicrw guardrails observability

Works forJev for Decision gating Jev for Verification

Used inJev for Software & DevTools Jev for Security & Privacy

Languages: TypeScript
Platforms: CLI
Capabilities: Open sourceRuns locally

Highlights

Block-or-allow verdicts happen in milliseconds with no model in the decision path, so verdicts are explainable and reproducible.
Claude Code and Codex are the supported adapters; Cursor, Copilot CLI, Gemini CLI, OpenCode and Pi are experimental.
Fails closed on destructive commands, secrets written into source and unvetted dependencies before they touch disk.
The author's 109-day audit log covers 3,834 guard_block events, of which 911 survived per-event verification against session transcripts.
State lives under .interlinked/ and the exit ramp is explicit: interlinked disable --uninstall removes a wizard installation.

Quickstart

bash

npm install -g github:QuentinCody/interlinked-cli
interlinked

Watch out

Copy reviewed against the repo README on 2026-10-15; the receipt numbers are the maintainer's own dogfooding on one machine, and the README says each experimental adapter can enforce only the controls its host API exposes.

Questions about this entry

Short answers drawn from the linked project or page.

Reactions & coverage

Posts, threads, and videos about this entry from around the web.

X: Agent Handoff Gate: agents verify what they hand off

X: Jev as an LLM guardrail inside Agentgateway

X: Vercel's fx safety reviewer, measured at 18× faster

Reddit: Early experiments wiring Jev into Pi as a safety layer and model router.

From the Jev guides

Original write-ups on what this entry is for.

Jev at home: voice pipelines, Assist routing, and sentinels

Oct 16, 2026

How Home Assistant conversation agents, macOS voice pipelines, and an advisory sentinel put a fast typed decision in the loop — one call per utterance, thresholds in code, and a daily token budget.

Jev Tools & CLIs#home-assistant#voice#automation

Read article

Search visibility, audited by decisions

Oct 16, 2026

Jev-based SEO tools split the crawl from the call: deterministic crawlers count and parse, while typed judgments rank fixes, place internal links, score search results, and track AI-answer mentions — with costs and caveats attached.

Jev Tools & CLIs#seo#search-visibility#audits

Read article

Jev on the Gateways: Vercel, Workers AI, and OpenRouter

Oct 12, 2026

The request shape barely changes across surfaces; the surrounding contract does. When to call the API directly and when a gateway, a Worker binding, or a router entry earns its extra hop.

Jev Tools & CLIs#gateway#vercel#cloudflare

Read article

From the Jev AI model community

Posts from builders shipping with Jev right now.

Follow @typesafeai

Interlinked CLI

Highlights

Quickstart

Questions about this entry

What is Interlinked CLI?

How does it differ from the agent's own permissions?

How do I install it?

Does it require a cloud service?

What do the audit numbers mean?

Does it work with every agent?

How do I turn it off?

Reactions & coverage

Related terms

More Jev resources like this

From the Jev guides

From the Jev AI model community