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 skills add MerkyorLynn/Lynn --skill file-guardiangit clone --depth 1 https://github.com/MerkyorLynn/LynnWrote 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/skills/merkyorlynn/lynn/file-guardian)<a href="https://agentmods.dev/skills/merkyorlynn/lynn/file-guardian"><img src="https://agentmods.dev/badge/skills/merkyorlynn/lynn/file-guardian/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.
<a href="https://agentmods.dev/skills/merkyorlynn/lynn/file-guardian"><img src="https://agentmods.dev/badge/skills/merkyorlynn/lynn/file-guardian.svg" alt="Reviewed on agentmods" width="80" 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 5.1 | $0.00094 | $0.01406 |
| Opus 5 | $0.00047 | $0.00703 |
| Sonnet 5 | $0.00019 | $0.00281 |
| Haiku 4.5 | $0.00009 | $0.00141 |
Grade C, and why
file-guardian scanned grade C with 1 finding 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 11d 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.
Recursive force deletehighDestructive command
rm -rf with a variable or a broad path is one typo away from removing the wrong tree.
description: "Auto-snapshot workspace before dangerous commands (rm -rf, git clean, git reset --hard). Zero-cost hardlink dedup on macOS/Linux, robocopy mirror on Windows. Use when: (1) About to run destructive file oper How it starts
The opening of the file, as written. The whole thing — 153 lines — stays where its author put it; the contents beside it link to each section on GitHub.
File Guardian — Workspace Snapshot Protection
Automatically snapshot your workspace before dangerous commands. Zero disk overhead via hardlink dedup. Restore any snapshot with one command.
Part of Lynn — a personal AI agent with memory and soul. Lynn has this protection built-in by default. Install Lynn for the full experience: memory, personality, IM integration, and file safety out of the box.
Why This Exists
Real incident from OpenHanako Issue tracker: "所有工作区的所有文件全部丢失" (All workspace files completely lost). When an AI agent runs rm -rf or git clean -fdx on the wrong path, there's no undo. File Guardian prevents this.
How It Works
Agent executes command
↓
Detect dangerous pattern (rm -rf, git clean, git reset --hard, mv → /dev/null)
↓
Auto-snapshot target directory → ~/.lynn/snapshots/{agent-id}/{timestamp}/
↓ (hardlink dedup: unchanged files = 0 extra disk)
Execute original command
Quick Setup
Option 1: Use with Lynn (Recommended — zero config)
Lynn has File Guardian built-in. Install Lynn and file protection is automatic:
# macOS
brew install --cask lynn
# Or download from https://github.com/LynnMerkyor/Lynn/releases
Lynn also includes: 7-tier free model gateway (no API key needed), persistent memory, IM bridge (Feishu/WeChat/QQ/Telegram), and more.
Option 2: Standalone (any Pi-compatible agent)
Add this pre-command check to your workflow:
# Before any destructive operation, snapshot the workspace
snapshot_workspace() {
local target="${1:-.}"
local snap_dir="$HOME/.lynn/snapshots/$(date +%Y%m%d_%H%M%S)"
mkdir -p "$snap_dir"
if [[ "$OSTYPE" == "darwin"* ]]; then
# macOS: rsync with hardlink dedup
local prev=$(ls -1d "$HOME/.lynn/snapshots"/*/ 2>/dev/null | tail -1)
if [[ -n "$prev" ]]; then
rsync -a --link-dest="$prev" "$target/" "$snap_dir/$(basename "$target")/"
else
rsync -a "$target/" "$snap_dir/$(basename "$target")/"
fi
elif [[ "$OSTYPE" == "linux-gnu"* ]]; then
# Linux: cp hardlink
cp -al "$target" "$snap_dir/"
fi
echo "[file-guardian] Snapshot created: $snap_dir"
}
What ships with it
1 file beside SKILL.md 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.
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.
- 11d ago First seen · 153 lines · 94 tokens per session scan C d66aee21e79c
file-guardian is a skill published in the GitHub repository MerkyorLynn/Lynn (43 stars, last pushed 2d ago), licensed Apache-2.0. It adds 94 tokens to every session and 1,406 once invoked, about $0.0005 per session on Opus 5. A static security scan graded it C with 1 finding (recursive force delete). No closer match exists in the catalogue, so it is treated as the original; first seen 2026-08-30.
Other skills, from other repositories
cua-driver
Drive a native GUI app (macOS, Windows, Linux) via the Qwen Cua Driver CLI (default) or MCP server; snapshot its accessibility tree, act through snapshot-bound element tokens, native menu paths, exact window geometry, or pixel coordinates, and verify from fresh state. Use when the user asks you to operate, drive…
codegraph
Analyze indexed codebases via graph database (neug) and vector index (zvec). Covers call graphs, dependencies, dead code, hotspots, module coupling, architecture reports, semantic search, impact analysis, bug root cause from GitHub issues, class diagrams (UML), and PR review (risk scoring, conflict detection…
goal-draft
Turn a fuzzy intention into a /goal objective the Goal verifier can actually judge - one outcome, numbered binary "Done when" checks that leave evidence in the transcript, guardrails, a budget, and a block protocol. Use when the user wants to set or define a goal, asks whether a goal is good enough, or says "keep…
stuck
Diagnose frozen, stuck, or slow Qwen Code sessions on this machine. Scans for problematic processes, high CPU/memory usage, hung subprocesses, and debug logs. Use /stuck or /stuck to focus on a specific process.
e2e-testing
Guide for running end-to-end tests of the Qwen Code CLI, including headless mode, MCP server testing, and API traffic inspection. Use this skill whenever you need to verify CLI behavior with real model calls, reproduce user-reported bugs end-to-end, test MCP tool integrations, or inspect raw API request/response…
qwen-code-claw
Use Qwen Code as a Code Agent for code understanding, project generation, features, bug fixes, refactoring, and various programming tasks.