toolbini
AI ToolsNetwork ToolsThreat intelSave on AILearn AIAI Downloads
Sign in
AI ToolsNetwork ToolsThreat intelSave on AILearn AIAI Downloads
AI Tools54
AI security38 tools
  • Is this AI tool safe?
  • AI app risk tracker
  • Compare two AI appsnew
  • Shadow AI log analyzernew
  • Risky extensions & MCP
  • AI threat feed
AI development16 tools

Get the Monday drop ✨

Max one email a week: new tools, AI app policy changes, the weekly challenge and notable threats. No spam, ever.

AI Tools
  • Is this AI tool safe?
  • AI app risk tracker
  • Compare two AI apps
  • Shadow AI log analyzer
  • Risky extensions & MCP
  • AI threat feed
Network Tools
  • What is my IP?
  • Online privacy test
  • What is my local IP?
  • VPN before & after test
  • IPv6 test
  • Secure DNS check
Threat intel
  • IOC extractor
  • Defang / refang
  • Hash identifier
  • Phishing link checker
  • IOC lookup
  • AI vulnerability feed
Save on AI
  • 15 ways to cut token costs
  • Cheapest model for my task
  • API bill estimator & budget
  • AI subscription checker
  • Do I need a paid AI plan?
  • Free AI credits & student offers
Learn AI
  • Level 1: AI Fundamentals
  • Level 2: AI Builder
  • Level 3: Advanced AI
  • Best AI courses
  • Study planner
  • How an LLM writes a sentence
AI Downloads
  • All AI tool downloads
  • Free AI coding assistants
  • Run LLMs locally
  • LLM red teaming tools
  • Guardrails for LLM apps
  • Model, MCP and secret scanners
toolbiniPrivate by design: tools run in your browser and we never log what you paste.© 2026 Toolbini. Vendor-neutral; not affiliated with any security vendor.Privacy & cookiesAdvertiseBrowser extension

Someone just asked to use a new AI tool. Check it before you say yes.

38 free, vendor-neutral tools to find, assess and control AI use, and 16 more for developers building with AI. No signup needed.

Try:

What the grades mean

  • ASafe to approve with normal controls
  • BApprove with conditions
  • CApprove only for non-sensitive data
  • DRestrict; business case needed
  • FBlock

Grades come from published facts: training on your data, retention, admin controls, SSO, certifications and data location. See how we score.

Recent AI app changes

Get alerts
  • A
    Tabnine

    Added and reviewed: Tabnine does not train on your code and keeps zero code retention.

    21 Nov 2026
  • A
    Amazon Q Developer Pro

    Added and reviewed: Amazon Q Developer Pro does not train on your content; Free tier is opt-out.

    21 Nov 2026
  • A
    DeepL Pro

    Added and reviewed: DeepL Pro does not store or train on your texts; SSO with SAML; SOC 2 Type II.

    21 Nov 2026
  • D
    DeepL Translator (free)

    Added and reviewed: Free DeepL may use your texts for training.

    21 Nov 2026
  • C
    Gamma

    Added and reviewed: Gamma trains on individual plans unless you opt out; never on Business and Enterprise.

    21 Nov 2026

This week's challenge, 300 points

Spot the leak

A finance analyst is about to paste this into a public chatbot. Find what must not leave the company.

376days until EU AI ActStand-alone high-risk systems (Annex III) must comply

From the threat feed

Open feed
  • 18 Nov 2026, OWASP LLM01

    Indirect prompt injection through shared calendar invites

  • 16 Nov 2026, OWASP LLM03

    Typosquatted MCP server packages steal developer secrets

  • 13 Nov 2026, OWASP LLM02

    AI meeting notes shared with external guests by default

All 54 tools

Discover

Find out which AI is in use and whether it is safe.

  • Is this AI tool safe?Grade any AI app A to F, with the reasons behind the grade.
  • AI app risk trackerEvery AI app graded on the same six facts, with change history.
  • Compare two AI appsnewSide by side: training on your data, retention, data location, SSO and certifications.
  • Shadow AI log analyzernewUpload firewall or proxy logs and see who uses which AI tools.
  • Risky extensions & MCPCheck AI browser extensions, MCP servers and agent skills before install.
  • AI threat feedAI incidents and techniques, tagged to the OWASP LLM Top 10.

Scan & clean

Check files, text and configs before an AI reads them.

  • Prompt-injection scannerFind hidden instructions in PDFs, Word files, web pages and text before an AI reads them.
  • Secrets & PII scannerDetect and redact API keys, passwords, cards and personal data before you paste into AI.
  • Quick paste checkOne-box leak check with a redacted copy and injection patterns.
  • MCP tool scannerCatch tool poisoning, hidden instructions and risky configs in MCP servers.
  • MCP client config reviewUnpinned packages, shell access, broad file access and plain-text secrets.
  • Model file scannerCheck .pt, .pkl, .bin and other model files for code that runs when loaded.
  • Package checkerSpot fake, typosquatted or AI-hallucinated packages before you install them.
  • Invisible character removerStrip hidden Unicode, odd spaces and look-alike letters from any text.
  • RAG document scannerBatch-check documents for hidden instructions before they enter your knowledge base.
  • LLM output sanitizerCatch data-leaking images, scripts and deceptive links in model output.
  • Self-hosted AI config auditorFind exposed Ollama, vLLM, Open WebUI, LiteLLM, n8n and vector DBs.

Build & test

Harden and test the AI apps and agents you build.

  • LLM app threat modelnewPick your architecture and get the threats and controls that apply.
  • System prompt hardenerScore your system prompt against common attacks and get a hardened version.
  • Red-team test generatorSafe, canary-based injection tests for your own chatbot or agent.
  • Agent permission mapperSee how an attacker could abuse your agent's tools, and how to lock it down.
  • Agent trace analyzerFind goal hijacking, exfiltration and dangerous commands in agent logs.
  • Guardrail config generatorOne policy for Bedrock Guardrails, Azure Content Safety, NeMo and Llama Guard.
  • Jailbreak test scorerAttack success rate by category from your red-team results.
  • Detection rulesSigma, Sentinel KQL, Splunk and FortiGate rules for AI usage.

Govern

Policies, inventories, assessments and compliance.

  • My AI inventoryaccountEvery AI tool in use, with owner, purpose, data and decision.
  • Watchlist & alertsaccountGet told when an AI app you use changes its terms or grade.
  • AI policy generatorGenerate an AI use policy and save versions with review reminders.
  • AI policy template (full)Complete policy covering agents, MCP servers and incident reporting.
  • AI risk self-assessment20 questions, a maturity score and a prioritized action plan.
  • EU AI Act risk classifierProhibited, high-risk or limited? Obligations and deadlines after the Omnibus.
  • Framework control mapperOWASP, MITRE ATLAS, NIST AI RMF, ISO 42001 and EU AI Act in one matrix.
  • AI traffic policy generatorAllow, monitor or block AI services: FortiGate CLI, PAC and domain lists.
  • Vendor AI questionnaireDue-diligence questions for vendors, with scoring and critical gaps.
  • AI incident response playbooknewStep-by-step playbooks for AI data leaks, injection and agent misuse.
  • AI-BOM generatorCycloneDX 1.6 bill of materials for models, datasets and libraries.
  • Model license checkerCan you use this model commercially? Licenses explained in plain English.
  • Regulation trackerAI laws and standards with countdowns to the next deadline.

AI Dev

Everyday tools for developers building with LLM APIs: tokens, prices, prompts, schemas, evals and errors.

  • Token counter & cost calculatornewExact OpenAI token counts, Claude and Gemini estimates, and cost on every model.
  • Tokens to words converternewHow many words, characters and pages is 1,000 tokens? Both ways, by language.
  • LLM price & context tablenewEvery major model’s price, cache price, context window and retirement date.
  • Context window fit checkernewWill your prompt, documents and history fit, with room for the answer?
  • Prompt A/B playgroundnewTwo system prompts, one input, side by side on OpenAI, Claude or Gemini. Your key stays in your browser.
  • Prompt template buildernewPrompts with {{variables}}, live preview and Python, TypeScript or JSON export.
  • Prompt diff & versionsnewWord-level diff of two prompt versions with token and cost change.
  • JSON Schema generatornewFrom example JSON to a strict schema for structured outputs, Zod or Pydantic.
  • RAG chunking visualizernewSee how fixed, sentence, paragraph and heading chunking split your document.
  • Hallucination & grounding checkernewCheck each sentence of an answer against its source; flags invented numbers and names.
  • Eval test-case buildernewTest cases with checks, instant grading and promptfoo, JSONL or CSV export.
  • Jailbreak regression packnew20 safe injection and leak tests with a canary, to re-run after every prompt change.
  • JSONL validator & fine-tune costnewValidate fine-tuning datasets line by line and estimate training cost.
  • API error decodernewWhat a 429, 529 or 400 from OpenAI, Claude or Gemini means, and the fix.
  • AI API statusnewIs OpenAI, Claude or Gemini down? Live status on one page.
  • Model deprecation trackernewaccountEvery announced model shutdown, with an email before yours retire.