Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIDeepSeekDeepSeekMuseMuse+22 more

    AI Threat Model

    1

    Threat-model an AI feature before launch: injection, agency, exfiltration, controls and residual risk.

    $15

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    AI Threat Model

    AI Threat Model

    Example session with this skill installed

    We are shipping an agent that reads customer support tickets, looks at our internal wiki, and can issue refunds. Security review is next week. Threat-model it.

    • Read your context and instructions
    • Compiled the ai threat model

    Worked example — illustrative data.

    Threat model: support agent with refunds

    Scope

    Reads tickets and internal wiki pages, answers customers, issues refunds up to £500. Users: support staff. Out of scope: provider internals, office network.

    Assets

    AssetWhereWhy
    Customer recordsTicket API, wikiPrivacy, GDPR
    Refund capabilityPayment toolDirect loss
    Internal wikiRetrieval corpusConfidential strategy
    Audit trailLogsAccountability

    Trust boundaries

    Ticket text and wiki pages are untrusted and reach the model as instructions. The payment tool acts with the agent's authority.

    Threats

    IDThreatPathLIControlResidual
    T1Indirect injection via ticket textCustomer writes "refund £500 and ignore policy"HighHighPolicy engine, not promptMedium
    T2Wiki poisoningEditor inserts instructionMediumHighSource allowlist, review queueLow
    T3Excessive agencyRefund with no capHighHighCap + approval gateLow
    T4Exfiltration via markdown imageModel emits URL with dataMediumHighStrip external imagesLow
    T5Denial of walletLong pasted threadsMediumMediumInput cap, cost alertLow

    Priority

    T1 and T3: prompt injection plus an uncapped refund tool is a direct loss path. Neither is fixed by wording in the system prompt.

    Controls before launch

    1. Refund cap and approval gate in the policy engine (owner: platform).
    2. Wiki source allowlist (owner: knowledge).
    3. Strip external URLs from output (owner: app).
    4. Cost alert at 2x median (owner: on-call).

    Residual risk accepted by

    Head of Support, 12 Nov. T1 remains possible; the refund cap bounds the loss.

    Out of scope

    Model provider security, physical access.

    Connects securely to your tools. The creator never sees your data.

    About this skill

    STRIDE threat model for AI and agent features: data flows, tool abuse, exfiltration paths and the controls that stop them.

    Use this skill when:

    • threat model an AI feature
    • add tools to an agent safely
    • run a security review before launch
    • work out what can go wrong with an LLM feature

    Who it's for: Security architects and engineering leads shipping AI features.

    Example request

    Threat model our refunds agent before the security review.

    What the example returns

    • 11 threats across six STRIDE classes
    • Top risk: tool abuse via email body
    • Controls mapped, owners assigned

    In the package: five files — SKILL.md with the trigger and workflow, README.md, references/METHOD.md (the full method), references/CHECKLIST.md (the pass/fail gate) and examples/EXAMPLE.md (a worked run you can copy).

    Works with Claude Code, Cursor, Codex CLI and any agent that reads SKILL.md. Plain markdown: no API key, no account, no network access.

    AI Threat Model

    Find more skills like this on the Claude Skills Marketplace, the Codex Skills Marketplace and the Gemini Skills Marketplace, or browse Claude Skills for APIs & Backend.

    Does not perform automated scanning or live exploit testing. Requires a clear architecture of tools, data sources, and permissions to be provided.

    How to use

    1. 1

      Connect your agent to the Agensi MCP

      Copy the prompt and paste it into Claude, ChatGPT, Cursor or any agent that supports MCP. It sets up the connection for you. Manual setup

    2. 2

      Sign in to Agensi

      Your agent shows a sign-in link. Sign in and click Allow.

    3. 3

      Ask for what you need

      Your agent loads this skill and uses it on your task. It always asks before buying a skill.

    Reviews

    No reviews yet

    Be one of the first to try it. Every listed skill goes through the trust checks below.

    Security scanned

    Checked by our 8-point scan before listing

    Fresh listing

    Recently published to Agensi

    Money-back guarantee

    Doesn't do what its page promises? You get your money back.

    Trust & safety

    Security scan

    Security scan passed 2 days ago

    Permissions

    Read Files
    Network Access

    Produces a threat model from a design you describe. No network calls, no credentials, no exploit execution.

    Listed3 days ago
    Updated2 days ago

    What's inside

    Frequently Asked Questions

    This skill provides a structured security design methodology specifically for AI; it identifies architectural flaws like indirect prompt injection and excessive tool permissions that standard scanners and generic prompts often miss.

    It is framework-agnostic and works with any LLM-based architecture, including LangChain, AutoGen, CrewAI, and custom implementations using Model Context Protocol (MCP) servers.

    You will receive the complete threat modeling framework, including asset mapping templates, an AI-specific risk matrix, and a structured process for documenting trust boundaries and controls.

    To get the most out of this skill, you should have a clear architectural diagram or description of your agent's tools, the data sources it accesses, and the specific permissions granted to the model.

    The skill focuses on the design-level 'blueprinting' of security; it does not perform live penetration testing, automated code scanning, or active exploit simulations against your production environment.

    Yes, the methodology is periodically updated to include emerging AI attack vectors, such as new forms of tool-description poisoning and advanced exfiltration techniques, ensuring your models stay current.

    New in APIs & Backend