A software-engineering method layer for Gemini CLI, a command-line coding assistant. It provides skills for debugging, planning, code review, handoffs, testing, releases, and coordinating multiple agents.

Use it to reproduce and fix bugs, compare approaches, critique plans, review code, coordinate agents, and check work before releases.

It gives the assistant a repeatable way to investigate problems and review decisions instead of relying on ad hoc steps. In Gemini CLI, it does not include the separate safety hooks or session memory described for other tools.

kernel — Plugin for Claude Code and Gemini CLI by ariaxhan

Install

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.

Clone the repo

git clone --depth 1 https://github.com/ariaxhan/kernel-claude

Made for: Claude Code, Gemini CLI.

Wrote 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.

README.md

[![agentmods](https://agentmods.dev/badge/plugins/ariaxhan/kernel-claude/gemini-extension/github.svg)](https://agentmods.dev/plugins/ariaxhan/kernel-claude/gemini-extension)

Your own site

<a href="https://agentmods.dev/plugins/ariaxhan/kernel-claude/gemini-extension"><img src="https://agentmods.dev/badge/plugins/ariaxhan/kernel-claude/gemini-extension/github.svg" alt="Measured on agentmods" height="20"></a>

Or the 80×15 button, for a site that already has a row of RSS and ATOM ones. Only the verdict fits; the numbers stay here.

Your own site · 80×15

<a href="https://agentmods.dev/plugins/ariaxhan/kernel-claude/gemini-extension"><img src="https://agentmods.dev/badge/plugins/ariaxhan/kernel-claude/gemini-extension.svg" alt="Reviewed on agentmods" width="80" height="20"></a>

Per session not measured What this adds to a session before it is invoked.

When invoked not measured Not applicable: nothing here is loaded into a session.

Security scan A 0 findings. A grade says what 26 rules found in the file — not that it is safe.

Origin original No closer match found in the catalogue.

Security

Grade A, and why

kernel scanned grade A with 0 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 1mo ago.

The scan reads gemini-extension.json. This mod also ships 14 executable files (scripts/adjudicate.py, scripts/build-gemini-extension.sh, scripts/bump-version.sh, …), listed below but not scanned — reading those needs a real analyzer, not pattern matching.

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.

Nothing flagged

None of the 26 patterns this scan looks for appear in this file: no shell pipes, no recursive deletes, no credential paths, no hidden text, no instruction-override or anti-refusal phrasing, no agent-config snooping. That is not a guarantee, it is the absence of the things that are checkable.

gemini-extension.json · 7 lines

What it actually says

{
  "name": "kernel",
  "version": "9.11.0",
  "description": "KERNEL's methodology layer for Gemini CLI: 27 software-engineering agent skills plus llms.txt as ambient context. Skills cover systematic debugging (reproduce, isolate, regression-test), generating competing approaches before implementing, pre-implementation critique, code review, bounded context handoff and resume, multi-agent orchestration, eval-driven development, release gating, and context-led frontend design. Gemini CLI gets the methodology only. KERNEL's enforcement layer does not run here: the PreToolUse hooks that block destructive commands and secret writes, the one-time human approval token for irreversible operations, and the agentdb SQLite memory recalled at session start are Claude Code and Codex only. `gemini extensions install` fetches a curated release bundle that carries just this manifest, llms.txt, and skills/, so no Claude-format hook or agent definition is loaded. Source, safety limits, and the host capability matrix: https://github.com/ariaxhan/kernel-claude",
  "contextFileName": "llms.txt"
}

Files

What ships with it

22 files beside gemini-extension.json in the same directory: the scripts, references and assets a skill reads on demand. Not counted in the per-session cost; read them before you install if any of them is executable.

Changes

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.

1mo ago Changed edb8f3a9e11d
1mo ago Changed 497ff9f5bebf
1mo ago First seen · 7 lines scan A 2241c862dd5a

Subscribe to this mod's changes

kernel is a plugin published in the GitHub repository ariaxhan/kernel-claude (14 stars, last pushed 2d ago), licensed MIT. Its token cost is not measured: this kind of file is read by the harness, not the model. A static security scan graded it A with 0 findings. No closer match exists in the catalogue, so it is treated as the original; first seen 2026-09-06.

Building agents Claude Code Debugging agent-memory ai-agents anthropic developer-tools llm-agents

causal-memory

Persistent causal memory for Claude Code — decisions, outcomes, and lessons that survive compaction. Same-context counterfactuals (natural experiments), a falsifiable prediction ledger, and an activation skill that teaches when to recall and record.

JingxuanC/causal-memory · not measured

distributed-debugging

Distributed system tracing and debugging across microservices.

HermeticOrmus/LibreUIUX-Claude-Code · not measured

yantrikdb

Cognitive memory for AI agents — semantic recall, a knowledge graph, contradiction detection and procedural learning. Local-first: one SQLite file, no account, no telemetry.

yantrikos/yantrikdb-mcp · not measured

aio-golang-mastery

Write and lint idiomatic Go end-to-end — a 7-tool lint chain (build, vet, golangci-lint, govulncheck, nilaway, deadcode, race) plus reference-grade patterns for concurrency, errors, generics, gRPC, naming/style enforcement (revive 21-rule set, golangci-lint v2 production config), and production hardening.

aiocean/claude-plugins · not measured

aio-ios-device-debug

Debug iOS apps on physical devices. Build, install, launch, capture logs, pull crash reports, and analyze crashes from the terminal.