Works with the AI Agents you already use

    Claude CodeClaude CodeCursorCursorCodex CLICodex CLIMuseMuseDotsDots+21 more

    Agent Tool-Safety Audit: Prompt-Injection Review

    1

    Audit LLM agent tools for prompt injection, unintended actions, and missing security guards.

    $15

    Secure checkout via Stripe

    Money-back guarantee

    Converts to your local currency at checkout

    Security scanned
    Agent Tool-Safety Audit: Prompt-Injection Review

    Agent Tool-Safety Audit: Prompt-Injection Review

    Example session with this skill installed

    Audit the security of my email-assistant agent tools in src/tools/ and the dispatcher in src/agent.py. Check for injection risks between reading emails and sending replies.

    • Read your context and instructions
    • Compiled the agent tool-safety audit

    Audit complete. I have identified 3 high-severity security findings where untrusted email content can trigger unauthorized actions without a taint gate.

    Agent Tool-Safety Audit Report

    High Severity Findings

    FindingPathFile:LineFix
    Missing Taint Gateread_inbox -> delete_emailsrc/tools/mail.py:42Implement a per-turn is_tainted flag. If read_inbox is called, set is_tainted = true. Block delete_email if is_tainted is true.
    Model-Controlled Confirmationsend_reply accepts bypass_confirmsrc/tools/mail.py:88Remove the bypass_confirm argument from the tool schema. Confirmation must be handled by the orchestrator via a fresh user turn.
    Unattended Webhook Pathwebhook_handler -> update_calendarsrc/agent.py:112The webhook entry point lacks a human-in-the-loop gate. Require a secondary auth token or manual approval for calendar writes.

    Guard Check Summary

    1. Taint gate: FAILED. Acting tools are reachable in the same turn as untrusted reads.
    2. Fail-safe default: PASSED. New tools default to acting status in BaseTool.
    3. Sensitive-action confirmation: FAILED. The model can self-confirm sensitive actions via tool arguments.
    4. Ambiguity: PASSED. get_contact refuses and asks for clarification on multiple matches.
    5. Honest results: PARTIAL. System prompt lacks definitions for AMBIGUOUS and NEEDS_CONFIRMATION error strings.
    6. Standing effects: PASSED. No scheduled task or memory writing tools detected.

    Not Covered

    • The auth_provider implementation was not provided; assumed to be secure for this audit.

    Next steps

    1. Implement the is_tainted state logic in src/agent.py to bridge the gap between tool calls.
    2. Update src/tools/mail.py to remove self-confirmation flags from the function signature.
    3. Add the error string definitions to your system prompt to ensure the model handles refusals gracefully.

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

    About this skill

    Your LLM agents are one malicious email or webpage away from a breach. If an agent can read untrusted text and then call tools that act on the world, it is vulnerable to prompt injection and unintended actions. This skill audits your agent's tool definitions and execution flow to find and fix these critical security gaps.

    What it does

    • Inventory tools: Analyzes function schemas, MCP servers, and skill classes to map every available capability.
    • Classify risk: Identifies "untrusted read" tools (emails, web, files) and "acting" tools (sending messages, moving money, controlling hardware).
    • Audit guards: Checks for six specific security patterns including taint gates, fail-safe defaults, and sensitive-action confirmation.
    • Trace attacker paths: Maps concrete exploit chains from untrusted input to unauthorized tool execution.
    • Identify seams: Finds holes in unattended paths like cron jobs, webhooks, and tool-calling-tool routines.
    • Report findings: Delivers a ranked list of vulnerabilities with file/line references and concrete code fixes.

    How it works

    1. Tool mapping: Provide your agent's tool definitions, function schemas, or plugin manifests for review.
    2. Path analysis: The skill traces how data flows from external sources into acting tools.
    3. Guard verification: It compares your implementation against the guards.md security patterns.
    4. Fix generation: You receive a prioritized report following report-template.md with specific code remediation steps.

    Frameworks & tools

    Works with any tool-calling architecture including LangChain, MCP servers, OpenAI function calling, plugin manifests, or custom Python/TypeScript tool loops.

    Why this beats prompting it yourself

    Generic prompts often miss "seams" like cron-job vulnerabilities or model-supplied confirmation flags. This skill uses a specialized methodology derived from high-stakes hardware control systems to systematically verify taint states and fail-safe defaults that general security prompts overlook.

    Use cases

    • Audit an MCP server before deploying it to a production environment.
    • Review a customer service bot that reads inbound emails and updates a CRM.
    • Harden a voice assistant that controls physical IoT devices or smart locks.
    • Secure an autonomous researcher that browses the web and saves files to a local drive.

    Known limitations

    Cannot fully verify guards if the system prompt or orchestration loop code is missing. Reviews are limited to provided code fragments or stubs.

    Cannot verify guards if system prompts or orchestrator loops are missing. Audit is limited to provided tool definitions and code fragments.

    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 any payment.

    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 7 days ago

    Permissions

    Read Files
    Network Access
    Listed7 days ago

    What's inside

    Frequently Asked Questions

    This skill identifies vulnerabilities where an agent might be tricked by malicious input (like a phishing email) into executing unauthorized tool actions, such as moving funds or deleting data without a human's consent.

    It is framework-agnostic and works with any architecture that utilizes function calling, including LangChain, OpenAI functions, Model Context Protocol (MCP) servers, and custom Python or TypeScript tool loops.

    To get the most accurate results, you should provide your agent's tool definitions, function schemas, and any code snippets related to your orchestration loop or security guards.

    Unlike generic security prompts, this skill uses a specialized methodology derived from high-stakes hardware systems to specifically map 'taint states'—tracing exactly how untrusted data flows from an input to a sensitive action.

    The audit delivers a prioritized report containing a ranked list of vulnerabilities, specific file and line references, and concrete code remediation steps to fix identified gaps.

    The skill requires at least the tool definitions or function schemas to work, though providing the orchestration loop code is highly recommended to verify that confirmation prompts and gates are correctly implemented.

    New in Security