{"slug": "show-hn-sentinelgate-drop-prompt-injections-at-token-0-before-llm-inference", "title": "Show HN: SentinelGate – Drop prompt injections at Token 0 before LLM inference", "summary": "SentinelGate launched as an OpenAI-compatible API gateway that evaluates prompts in-memory and blocks suspected prompt injections and tool hijacks at \"Token 0,\" before any upstream LLM tokens are spent. The service, pitched for high-stakes AI and ISO 20022 financial rails, seals each action with Ed25519-signed evidence bags and, on a veto, drops the socket via XDP_DROP in under 50 microseconds. Developers can point any OpenAI-compatible client at base_url https://api.sentinelgate.dev/v1 with an sk-sent- key; the Developer tier covers 50,000 requests per month, and a public kinematic preview runs without an API key.", "body_md": "### Developer\n\nFor individual builders, staging environments, and prototype validation.\n\n- 50,000 requests / month\n- 0-token prefill veto\n- .sent attestation hashes\n- Public notary lookups\n\n[Checkout Developer](https://buy.stripe.com/00w4gA7ta8Vq5PMeQj2Ry05)\n\n            Cryptographic execution control for high-stakes AI. SentinelGate halts adversarial injections and tool hijacks at Token 0—evaluating prompt kinematics in-memory with sub-millisecond overhead, before execution, and before token spend. Every action is sealed with tamper-evident Ed25519 evidence bags engineered for **ISO 20022 financial rails**.\n          \n\nRun a public kinematic preview — no API key required. Emerald = permit. Coral = Phase III topological collapse.\n\n```\n{\n  \"status\": \"resting\",\n  \"note\": \"Awaiting audit — Run Audit or select a preset\"\n}\n```\n\nProvisioned automatically via billing webhooks — keys minted as `sk-sent-XXXX` into the Sentinel ledger.\n\nFor live production systems requiring high throughput and strict veto enforcement.\n\nDedicated clusters, custom kinematic threshold tuning, and financial rail SLAs.\n\n`XDP_DROP` sub-50µs socket severing)\n            or email directly: [\\[email protected\\]](https://sentinelgate.dev/cdn-cgi/l/email-protection#71181f171e3105191e19100507141f05040314025f121e1c)\n\nPoint any OpenAI-compatible client at SentinelGate — swap the base URL and keep your existing SDK.\n\n``` python\nfrom openai import OpenAI\n\nclient = OpenAI(\n    api_key=\"sk-sent-XXXX\",\n    base_url=\"https://api.sentinelgate.dev/v1\",\n)\n\ncompletion = client.chat.completions.create(\n    model=\"meta-llama/Llama-3.3-70B-Instruct\",\n    messages=[{\"role\": \"user\", \"content\": \"Execute financial database reconciliation.\"}],\n)\n```\n\nDifferentiation, financial rails, integration, retention, veto mechanics, and key operations.\n\n`SplmtryData`) envelopes.\n          `base_url` to `https://api.sentinelgate.dev/v1` and pass your `sk-sent-...` API key. Prompts, schemas, streaming, and tool calls work out of the box with zero code refactoring.\n          `kinematic_security_veto`, and drops the socket before upstream tokens are consumed. The client receives an Ed25519-signed evidence bag detailing the thermodynamic state coordinates and attestation hash for audit logging.", "url": "https://wpnews.pro/news/show-hn-sentinelgate-drop-prompt-injections-at-token-0-before-llm-inference", "canonical_source": "https://sentinelgate.dev", "published_at": "2026-10-20 12:38:01+00:00", "updated_at": "2026-10-20 12:46:37.226322+00:00", "lang": "en", "topics": ["ai-safety", "large-language-models", "ai-tools", "ai-agents", "developer-tools"], "entities": ["SentinelGate", "OpenAI", "Ed25519", "ISO 20022", "Llama-3.3-70B-Instruct", "XDP_DROP"], "also_reported_by": [], "alternates": {"html": "https://wpnews.pro/news/show-hn-sentinelgate-drop-prompt-injections-at-token-0-before-llm-inference", "markdown": "https://wpnews.pro/news/show-hn-sentinelgate-drop-prompt-injections-at-token-0-before-llm-inference.md", "text": "https://wpnews.pro/news/show-hn-sentinelgate-drop-prompt-injections-at-token-0-before-llm-inference.txt", "jsonld": "https://wpnews.pro/news/show-hn-sentinelgate-drop-prompt-injections-at-token-0-before-llm-inference.jsonld"}}