Agent Skills for integrating k-ID — age-appropriate compliance as a service — into your app, game, or platform. These skills cover the full range of jurisdictional regimes a k-ID integration can be configured for:
- COPPA (US)
- GDPR-Kids (EU)
- UK AADC and the UK Online Safety Act (highly effective age assurance for 18+ features)
- Brazil ECA Digital (loot-boxes and targeted ads at 18, direct marketing at 12)
- Australia Online Safety Act (social media minimum age)
- Other regional regimes configured in k-ID Compliance Studio
They follow the open Agent Skills specification and work with 35+ AI coding agents including Claude Code, Cursor, OpenAI Codex, GitHub Copilot, Gemini CLI, and others.
Canonical reference for API shapes and endpoints remains
docs.k-id.com. These skills encode integration
patterns, compliance invariants, and known pitfalls — not API documentation.
The skills cover both k-ID integration shapes. Use whichever matches your product.
- Shape A — Full sessioned integration. Age gate → session → (consent | verification | threshold) → permissions. For games, social platforms, and multi-feature apps that need persistent per-user state and multiple gated features.
- Shape B — Standalone AgeKit+. One call to the age-verification endpoint, no session, no permission-management UI. For 18+ sites (UK OSA), age-restricted downloads, AU social-media minimum-age checks, or any single age-proof decision.
The router skill detects which shape applies from the user's request.
Within Shape A, there is a second choice: custom UX workflows (build the UI, call the APIs) or widgets (pre-built k-ID iframes).
- Custom UX workflows (the default). Build the age gate and
consent UI yourself and call
/age-gate/check,/challenge/send-email, etc. directly. Produces the best-looking, most brand-integrated experience, renders inline (no iframe), and works on every platform k-ID supports — including Unity WebGL, consoles, and native desktop apps where iframes aren't practical. Recommended for production integrations. - Widgets (fast-path fallback). k-ID publishes drop-in widgets
for the age gate, the end-to-end flow (age gate + parental consent
- data notices + permissions + preferences), manage-permissions
upgrades, and data notices. They carry built-in
jurisdiction-appropriate age collection, initiate parental-consent
challenges automatically, and emit DOM events
(
Widget.AgeGate.Result,Widget.AgeGate.Challenge) to the parent window. Choose when the integration must be small, simple, and as fast as possible to ship — proofs of concept, internal tools, early-stage games.
- data notices + permissions + preferences), manage-permissions
upgrades, and data notices. They carry built-in
jurisdiction-appropriate age collection, initiate parental-consent
challenges automatically, and emit DOM events
(
The k-id-age-gate and k-id-consent-and-challenges skills cover
both approaches — the router picks the right pattern from the user's
intent. See
Choose integration
for the canonical comparison.
Eight skills, designed to work together:
| Skill | Purpose |
|---|---|
k-id-integration |
Router. Picks the right shape and sibling skill(s) based on the user's request. |
k-id-age-gate |
Entry point for Shape A — /age-gate/check, jurisdiction detection, correct age-appropriate defaults. |
k-id-consent-and-challenges |
Shape A only — GUARDIAN-managed parental consent flow for minors (QR + OTP + email) and challenge polling. |
k-id-age-verification |
Age verification and assurance for both shapes — standalone AgeKit+ (Pattern 1), unverified-adult in a session (Pattern 2), and per-permission threshold (Pattern 3, UK OSA 18+, Brazil ECA Digital, Australia social media age). |
k-id-sessions-and-permissions |
Shape A only — live permission propagation, /session/upgrade, permission-gated UI controls, verifiedAgeThreshold handling. |
k-id-webhooks |
Webhook signature verification and event handling. Relevant to both shapes. |
k-id-server-trust-boundary |
Companion skill: API key placement, server proxy, pre-flight checks. Required for both shapes. |
k-id-mobile-native |
Companion skill: iOS/Android/Unity platform age signals and deep links. |
A separate family of skills that orchestrate the Neimo MCP rather than the k-ID API. Use these when the agent needs to research children's-digital-safety regulation across multiple markets (COPPA, GDPR Art. 8, UK OSA, Brazil ECA Digital, AU social-media min-age, etc.) before or after touching k-ID itself.
| Skill | Purpose |
|---|---|
neimo-compliance-audit |
Multi-market compliance audit. Orchestrates Neimo's lookup_regulation, compare_regulations, lookup_state_laws, search_kb_semantic, search_primary_sources, lookup_market_insights to produce a structured findings document with citations and a tiered must / should / consider obligations table. |
Prerequisite: a Neimo MCP connection. Get a free trial key at neimo.k-id.com/connect. Without it the agent will report "Neimo MCP not configured" and stop.
A separate family of skills that produce Neimo-styled HTML deliverables rather than calling the k-ID API or the Neimo MCP. Use these when the agent is asked to build a visual asset (client blueprint, data story, presentation deck, blog diagram, or marketing mock).
| Skill | Purpose |
|---|---|
neimo-design-system |
Neimo's visual design system for HTML deliverables. Defines the shared design DNA (grain, typography, color semantics, components) and five asset modes (Client Blueprint, Data Story, Presentation Deck, Blog Diagram, Marketing Mock) with their tokens, layouts, and component libraries. Activates on "data story", "client blueprint", "Neimo look", and similar visual-asset requests. |
/plugin marketplace add https://github.com/kidentify/skills
/plugin install k-id-skills@kidentifyFor just the neimo. MCP and design skills (skip the k-ID integration bundle), swap the install line for:
/plugin install neimo-skills@kidentifyClone into your project's .agents/skills/ directory:
mkdir -p .agents/skills
git clone --depth 1 https://github.com/kidentify/skills.git .agents/k-id-skills-tmp
cp -r .agents/k-id-skills-tmp/skills/* .agents/skills/
rm -rf .agents/k-id-skills-tmpFor tool-specific GUI install paths, slash commands, and verification steps, see the full install matrix on docs.k-id.com/get-started/agent-skills.
Download the repo as a ZIP via the green "Code" button on GitHub, extract,
and copy the skills/ contents into your tool's skills directory.
The k-id-integration router is the first skill that activates on any k-ID
integration request. Based on the user's question, it points the agent at
the right feature skill (age gate, consent, verification, etc.), which in
turn may pull in companion skills (trust boundary, mobile native).
Each skill is self-contained: a SKILL.md with invariants and gotchas
inline, plus focused reference material in references/ when the agent
needs more detail on demand.
- Doc-first. API request/response shapes come from
docs.k-id.com. Skills encode patterns and pitfalls, not endpoints. - Invariants inline, reference on demand. The highest-value content —
the rules that prevent real, shipped bugs — lives directly in each
SKILL.md. Detailed reference material is loaded only when needed. - Calibrated prescriptiveness. Exact commands for fragile operations (API bodies, signature verification). Higher freedom for UI adaptation.
- Defaults, not menus. One recommended path per choice point; escape hatches noted where they matter.
- Validate your work. Every skill ends with a verification step so the agent confirms the integration is actually working.
This repository is closed to external contributions. To report a bug or
request a correction, open an issue on GitHub or email
developers@k-id.com.
Licensed under the Apache License 2.0 — see LICENSE.
Additional acceptable-use terms apply: these skills may not be used or
modified to circumvent age verification, parental consent, or other
child-protection controls, and your use of the k-ID API and hosted services
remains governed by your MSA. See NOTICE for the full
acceptable-use policy and additional terms.
- k-ID Developer Hub — canonical API reference
- Agent Skills specification — format these skills follow
- Agent Skills client showcase — compatible AI tools