skarn

A security plugin collection for AI coding assistants, including guards for Claude Code and Codex CLI and a tool for reconstructing work from coding sessions. The guards check for exposed secrets and data exfiltration in audit mode.

In plain words
What is it for?
Auditing Claude Code or Codex CLI actions, checking assistant configurations, and reconstructing recent work or statistics from AI coding sessions with redacted exports.
Why use it?
It helps review assistant activity and configuration for security risks. Separate pieces cover real-time checks and reading past sessions, so you can use the part relevant to the task.

Plugin for Claude Code

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.

Claude Code
/plugin marketplace add skarn-security/agent-guard
agentmods
npx agentmods add plugins/skarn-security/agent-guard/marketplace
Clone the repo
git clone --depth 1 https://github.com/skarn-security/agent-guard

Made for: Claude Code.

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. Scan, not verified.
Origin original No closer match found in the catalogue.
Security

Grade A, and why

skarn 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 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.

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.

.claude-plugin/marketplace.json · 45 lines

What it actually says

{
  "$schema": "https://anthropic.com/claude-code/marketplace.schema.json",
  "name": "skarn",
  "description": "Skarn security guard for AI coding assistants",
  "owner": {
    "name": "Skarn",
    "url": "https://getskarn.com"
  },
  "plugins": [
    {
      "name": "skarn-guard",
      "description": "Real-time pre-execution secret and exfiltration guard for Claude Code (audit mode), with the skarn-audit skill.",
      "author": {
        "name": "Skarn",
        "url": "https://getskarn.com"
      },
      "category": "security",
      "source": "./claude",
      "homepage": "https://getskarn.com"
    },
    {
      "name": "skarn-guard-codex",
      "description": "Real-time pre-execution secret and exfiltration guard for Codex CLI (audit mode), with the skarn-audit skill.",
      "author": {
        "name": "Skarn",
        "url": "https://getskarn.com"
      },
      "category": "security",
      "source": "./codex",
      "homepage": "https://getskarn.com"
    },
    {
      "name": "skarn-recall",
      "description": "Reconstruct what you worked on from your AI coding sessions: recent, stats, and redacted export. Reads session content, so it ships separately from the guard.",
      "author": {
        "name": "Skarn",
        "url": "https://getskarn.com"
      },
      "category": "productivity",
      "source": "./recall",
      "homepage": "https://getskarn.com"
    }
  ]
}
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.

  1. 2d ago First seen · 45 lines scan A 31b4bcf66588

Subscribe to this mod's changes

skarn is a plugin published in the GitHub repository skarn-security/agent-guard (31 stars, last pushed 5d 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 2026-08-30.