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 zscaler/zguard-ai-integrations --skill claude-code-skillgit clone --depth 1 https://github.com/zscaler/zguard-ai-integrationsWrote 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/zscaler/zguard-ai-integrations/claude-code-skill)<a href="https://agentmods.dev/skills/zscaler/zguard-ai-integrations/claude-code-skill"><img src="https://agentmods.dev/badge/skills/zscaler/zguard-ai-integrations/claude-code-skill.svg" alt="Measured on agentmods" 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 | $0.00230 | $0.00852 |
| Opus 5 | $0.00115 | $0.00426 |
| Sonnet 5 | $0.00046 | $0.00170 |
| Haiku 4.5 | $0.00023 | $0.00085 |
Grade A, and why
aiguard 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 6d 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.
How it starts
The opening of the file, as written. The whole thing — 95 lines — stays where its author put it; the contents beside it link to each section on GitHub.
Zscaler AI Guard Security Scanner
Detect security threats in prompts, AI responses, and code using Zscaler AI Guard.
What It Detects
All detectors enabled in your AI Guard policy are applied automatically. Available detectors include:
Security: Prompt Injection, Code, Secrets, PII, Personal Data, PII DeepScan, Invisible Text, Malicious URL Toxicity: Toxicity (multilingual) General: Text (regex), Gibberish, Competition, Language, Brand and Reputation Risk, Refusal, URL Reachability, Prompt/Response Tags Content Moderation: Legal Advice, Intellectual Property, Off Topic, Topic, Finance Advice
Prerequisites
Environment variables required:
AIGUARD_API_KEY— API key from AI Guard Console (Private AI Apps → App API Keys)AIGUARD_CLOUD— Cloud region (us1, us2, eu1, eu2)AIGUARD_POLICY_ID— (Optional) Specific policy ID; auto-resolved if not provided
How to Pass Content
Method 1: Heredoc (recommended — handles quotes and newlines)
python3 scripts/scan.py --type prompt <<'EOF'
Content with "quotes" and
multiple lines works fine.
EOF
Method 2: File (recommended for code)
python3 scripts/scan.py --type code --file path/to/file.py
Method 3: Direct argument (simple content only)
python3 scripts/scan.py --type prompt --content "simple text"
Conversation (prompt + response together)
python3 scripts/scan.py --type conversation --prompt "user prompt" --response "ai response"
Scan Types
| Type | Use Case |
|---|---|
prompt |
Scan user input before processing (direction: IN) |
response |
Scan AI-generated content before delivery (direction: OUT) |
code |
Scan generated code for vulnerabilities (direction: OUT) |
conversation |
Scan both prompt and response together |
Interpreting Results
- action: ALLOW — Content is safe, proceed normally
- action: DETECT — Threat detected but not blocked (alert); review and decide
- action: BLOCK — Threat detected and blocked; do NOT proceed with this content
What ships with it
5 files 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.
- 6d ago First seen · 95 lines · 230 tokens per session scan A ee095000288b
aiguard is a skill published in the GitHub repository zscaler/zguard-ai-integrations (5 stars, last pushed 4d ago), licensed MIT. It adds 230 tokens to every session and 852 once invoked, about $0.0011 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 2026-08-31.
Other skills, from other repositories
implementing-zero-trust-network-access-with-zscaler
Use when implement Zero Trust Network Access using Zscaler Private Access (ZPA) to replace traditional VPN with identity-based, context-aware access to private applications through the Zscaler Zero Trust Exchange. Use when implementing zero trust network access using zscaler private access (zpa).
configuring-zscaler-private-access-for-ztna
Configuring Zscaler Private Access (ZPA) to replace traditional VPN with zero trust network access by deploying App Connectors, defining application segments, configuring access policies based on user identity and device posture, and integrating with IdPs.
configuring-zscaler-private-access-for-ztna
Configures Zscaler Private Access (ZPA) to replace traditional VPN with zero trust network access by deploying App Connectors, defining application segments, configuring identity- and device-posture-based access policies, and integrating with IdPs. Use when replacing VPN concentrators with ZTNA or providing remote…
nemo-guardrails
NVIDIA's runtime safety framework for LLM applications. Features jailbreak detection, input/output validation, fact-checking, hallucination detection, PII filtering, toxicity detection. Uses Colang 2.0 DSL for programmable rails. Production-ready, runs on T4 GPU.
implementing-secrets-management-with-vault
This skill covers deploying HashiCorp Vault for centralized secrets management across cloud environments, including dynamic secret generation for databases and cloud providers, transit encryption, PKI certificate management, and Kubernetes integration. It addresses eliminating hardcoded credentials from application…
token-tracker
Track token consumption, enforce session budgets, and display cost for every NetClaw interaction.