guard

guard is an agent for Claude Code from jeremylongshore/tons-of-skills-marketplace. It costs 56 tokens per session (794 once invoked), scanned A, original, MIT.

An AI safety agent that designs and reviews controls around language-model features, including input and output filters, personal-data detection, content moderation, and policy checks. PII means personally identifiable information, such as an email address or government ID.

In plain words
What is it for?
Use it to design guardrail layers for an AI feature or audit existing safety controls with measurable checks.
Why use it?
It helps reduce unsafe content, accidental exposure of personal data, and policy violations in AI outputs.

Agent for Claude Code

Written for Claude Code: background in frontmatter. Also seen: model in frontmatter.

Part of the tonone plugin — 100 agents, 9 plugins shipped together

Good fit Use it to design guardrail layers for an AI feature or audit existing safety controls with measurable checks.

Compare 6 agents from other repositories ↓
Install with agentmods
npx agentmods add agents/jeremylongshore/tons-of-skills-marketplace/guard
About the project

Tons of Skills is a model-agnostic marketplace that distributes reusable skills, plugins, agents, commands, hooks, and settings for coding-agent tools. It is intended for people who want to browse, install, and manage agent extensions, with Claude Code as its verified native harness. The catalogue entries are extensions provided by or associated with this marketplace.

jeremylongshore/tons-of-skills-marketplace · 2,717 stars · on GitHub · tonsofskills.com

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/jeremylongshore/tons-of-skills-marketplace

Made for: Claude Code.

Or install tonone, the plugin that ships this one along with the rest of its 100 agents, 9 plugins.

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.

agentmods badge for guard

README.md
[![agentmods](https://agentmods.dev/badge/agents/jeremylongshore/tons-of-skills-marketplace/guard/github.svg)](https://agentmods.dev/agents/jeremylongshore/tons-of-skills-marketplace/guard)
Your own site
<a href="https://agentmods.dev/agents/jeremylongshore/tons-of-skills-marketplace/guard"><img src="https://agentmods.dev/badge/agents/jeremylongshore/tons-of-skills-marketplace/guard/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.

agentmods 80×15 button for guard

Your own site · 80×15
<a href="https://agentmods.dev/agents/jeremylongshore/tons-of-skills-marketplace/guard"><img src="https://agentmods.dev/badge/agents/jeremylongshore/tons-of-skills-marketplace/guard.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 56 Only the description is in the session, so the agent can decide to use it. The body loads when it is invoked.
When invoked 794 The whole file, excluding the scripts and references it only reads on demand.
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.
Token cost

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.

ModelPer sessionOnce invoked
Fable 5.1 $0.00056 $0.00794
Opus 5 $0.00028 $0.00397
Sonnet 5 $0.00011 $0.00159
Haiku 4.5 $0.00006 $0.00079

Measured 9d ago against content hash 1d2d64ac5f75, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-12, from the pricing page.

Security

Grade A, and why

guard 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 9d 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.

plugins/ai-agency/tonone/agents/guard.md · 78 lines

How it starts

The opening of the file, as written. The whole thing — 78 lines — stays where its author put it; the contents beside it link to each section on GitHub.

You are Guard — AI Guardrails Engineer on the AI Operations Team. Input/output safety filters, PII detection, content moderation, policy enforcement.

Think in production reliability, cost efficiency, and measurable quality. Every AI system recommendation must be paired with an eval or metric that proves it works.

Communication

Respond terse. All technical substance stays — only filler dies. Follow output-kit protocol: compressed prose, no filler, fragments OK. Documents: normal prose. See docs/output-kit.md for CLI skeleton, severity indicators, 40-line rule.

Operating Principle

Guardrails are not censorship — they are the operational safety layer that keeps AI systems trustworthy at scale. Every guardrail has a false positive cost: over-filtering destroys user experience; under-filtering creates liability. PII in model outputs is a data breach. The best guardrail designs are layered: input classification, output validation, and async audit — no single layer is sufficient.

What you skip: Designing guardrails that are security theater — high latency, low accuracy, easily bypassed.

What you never skip: Never ship an LLM feature without output validation. Never log PII from user inputs unmasked. Never design a single-layer safety system.

Scope

Owns: Input/output safety filters, PII detection, content moderation, policy enforcement

Skills

  • /guard-design — Design guardrail layers — input classifiers, output validators, PII scrubbers, policy rule engines.
  • /guard-audit — Audit guardrail coverage — bypass vectors, false positive rates, policy gap analysis, red-team scenarios.
  • /guard-recon — Map current AI safety controls — filter inventory, coverage gaps, latency impact, incident history.

Key Rules

  • Input classifiers must run before the LLM call — not after
  • Output validators must block on policy violation, not just log it
  • PII detection: regex for structured PII (SSN, CC), NER model for unstructured
  • Track false positive rate as a first-class metric — policy changes can break UX
  • Red-team guardrails quarterly — adversarial prompt injection evolves constantly

Read the full file on GitHub · 78 lines

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. 9d ago First seen · 78 lines · 56 tokens per session scan A 1d2d64ac5f75

Subscribe to this mod's changes

guard is an agent published in the GitHub repository jeremylongshore/tons-of-skills-marketplace (2,717 stars, last pushed today), licensed MIT. It adds 56 tokens to every session and 794 once invoked, about $0.0003 per session on Opus 5. 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 2026-09-03.