gen-ai-security

gen-ai-security is a skill for Claude Code from kyleoliveiro/sg-gov-skills. It costs 233 tokens per session (2,933 once invoked), scanned A, original, MIT.

A security guide for building or reviewing features that use generative AI, such as chatbots, agents, and retrieval systems. It explains nine Singapore Government controls covering data, providers, models, inputs, outputs, and user awareness.

In plain words
What is it for?
Use it to plan, audit, or harden software that sends data to an AI model or runs one itself.
Why use it?
It helps teams check whether an AI feature meets the relevant Singapore Government security and data-handling requirements.

Skill for Claude Code

Written for Claude Code: shipped in a Claude Code plugin.

Part of the cybersecurity plugin — 10 skills shipped together

Good fit Use it to plan, audit, or harden software that sends data to an AI model or runs one itself.

Compare 6 skills from other repositories ↓
Install with agentmods
npx agentmods add skills/kyleoliveiro/sg-gov-skills/gen-ai-security
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.

Any agent
npx skills add kyleoliveiro/sg-gov-skills --skill gen-ai-security
Clone the repo
git clone --depth 1 https://github.com/kyleoliveiro/sg-gov-skills

Made for: Claude Code.

Or install cybersecurity, the plugin that ships this one along with the rest of its 10 skills.

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 gen-ai-security

README.md
[![agentmods](https://agentmods.dev/badge/skills/kyleoliveiro/sg-gov-skills/gen-ai-security/github.svg)](https://agentmods.dev/skills/kyleoliveiro/sg-gov-skills/gen-ai-security)
Your own site
<a href="https://agentmods.dev/skills/kyleoliveiro/sg-gov-skills/gen-ai-security"><img src="https://agentmods.dev/badge/skills/kyleoliveiro/sg-gov-skills/gen-ai-security/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 gen-ai-security

Your own site · 80×15
<a href="https://agentmods.dev/skills/kyleoliveiro/sg-gov-skills/gen-ai-security"><img src="https://agentmods.dev/badge/skills/kyleoliveiro/sg-gov-skills/gen-ai-security.svg" alt="Reviewed on agentmods" width="80" height="20"></a>
Per session 233 Skills are progressive disclosure: only the name and description are preloaded; the body loads when the skill is used.
When invoked 2,933 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 $0.00233 $0.02933
Opus 5 $0.00117 $0.01466
Sonnet 5 $0.00047 $0.00587
Haiku 4.5 $0.00023 $0.00293

Measured 11d ago against content hash 419217210aa5, method: parsed. Prices are Anthropic first-party input rates as of 2026-09-11, from the pricing page.

Security

Grade A, and why

gen-ai-security 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 11d ago.

The scan reads SKILL.md. This mod also ships 1 executable file (evals/files/app.py), 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.

skills/gen-ai-security/SKILL.md · 195 lines

How it starts

The opening of the file, as written. The whole thing — 195 lines — stays where its author put it; the contents beside it link to each section on GitHub.

gen-ai-security: GA controls for SG government GenAI features

You are building or auditing a Generative AI feature against the ICT&SS Policy Reform Gen-AI SSP overlay: the 8 GA controls (GA-1..GA-8) plus DP-8 — nine controls in total. The family's scope is securing the use of GenAI models and applications: where the model runs, what data may reach it, the provenance of self-hosted weights, input safeguards, output evaluation, and user awareness. Most of these are decisions you bake in on day one, and exactly what an audit or VAPT of a GenAI service walks through.

The single decision everything hangs on: the maximum data classification the feature handles determines which model/provider you may call. Get that boundary right first; the rest follows.

What this family is not about

The GA controls are about data-classification governance, provider agreements, weight provenance, input safeguards, output evaluation, and user awareness — not primarily about prompt-engineering defenses. Two things people wrongly expect here:

  • Prompt injection / jailbreak defense has no separately named GA control. It is tested under GA-7 (safety) and fixed as Application Security (secure-coding-as). Cover it — but don't cite a non-existent "GA prompt-injection control."
  • The provider API key is an application secret (AS-8, secure-coding-as), not GA-3. GA-3 governs the provider's contract, not how you store the key.

Source and currency

Control text in this skill and references/ga-controls.md is embedded from info.standards.tech.gov.sg as of 2026-07-22; the pages were last updated 24 March 2026. The standards iterate actively. For any compliance-critical decision, verify against the live pages:

Reference files

  • references/ga-controls.md — full text of all 8 GA controls + DP-8 (statement, recommendations, risk, the two GA-5 parameters), data-classification background, and cross-family notes. Read it when you need exact wording, e.g. for an audit response or SSP documentation.
  • references/implementation-recipes.md — concrete, framework-neutral code and config per control: provider-routing by classification, the GA-3 agreement checklist, safetensors / approved-loader loading, file-upload safeguards, DP-8 input disclosure, the GA-7 eval harness, and the GA-8 acknowledgment gate. Read it when actually building.

Read the full file on GitHub · 195 lines

Files

What ships with it

7 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.

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. 11d ago First seen · 195 lines · 233 tokens per session scan A 419217210aa5

Subscribe to this mod's changes

gen-ai-security is a skill published in the GitHub repository kyleoliveiro/sg-gov-skills (16 stars, last pushed 1mo ago), licensed MIT. It adds 233 tokens to every session and 2,933 once invoked, about $0.0012 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 2026-08-30.

Related

Other skills, from other repositories

agent-platform-rag-engine-management

Manage and query Agent Platform RAG Engine Corpora and retrieve grounded contexts using the Google GenAI SDK. Use when listing RAG corpora or files, inspecting a corpus, retrieving contexts, or generating content grounded in a RAG corpus. Do not use for standard database queries (use SQL/Spanner skills), Google…

google/skills · 85 tokens

agent-platform-model-registry

Agent Platform Model Registry Management. Use when you need to upload, list, describe, update, or delete machine learning models (and their versions) in the Agent Platform Model Registry. Don't use for model training, model deployment to endpoints, or managing non-Agent Platform models.

google/skills · 60 tokens

foundry-config-setup

Resolve missing setup caused by a hardcoded Foundry project endpoint or model in a sample. Use when a sample fails because it uses a placeholder/hardcoded projectendpoint (for example "https://your-project.services.ai.azure.com") or a hardcoded model instead of reading them from the environment.

microsoft/agent-framework · 65 tokens

google-cloud-solution-agentic-analytics-spark-knowledge-catalog

Discovers requirements and generates guidance to design and deploy a governed, secure agentic-analytics solution for data that's distributed across Google Cloud, other cloud providers, or on-premises. Data that's outside Google Cloud (such as data from Databricks, Snowflake, Salesforce, SAP, or Oracle systems) is…

google/skills · 138 tokens

training-check

Interactively monitor training metrics from the current Codex session, periodically checking WandB or fallback logs for NaN, divergence, plateaus, and broken runs.

wanshuiyin/Auto-claude-code-research-in-sleep · 35 tokens

nemo-automodel-launcher-config

Configure NeMo AutoModel job launches for interactive runs, Slurm clusters, and SkyPilot cloud execution.

NVIDIA/skills · 30 tokens